From 0a3b7332775f97edd40f6f0b3684d4a560fca7d0 Mon Sep 17 00:00:00 2001
From: "hongliang.yuan" <hongliang.yuan@iluvatar.com>
Date: Wed, 26 Mar 2025 13:13:54 +0800
Subject: [PATCH] add supported env for ci not ready models

---
 .../conformer/paddlepaddle/README.md          |   6 +
 .../conformer_wenet/pytorch/README.md         |   6 +
 .../pytorch/README.md                         |   6 +
 .../speech_recognition/rnnt/pytorch/README.md |   6 +
 .../transformer_wenet/pytorch/README.md       |   6 +
 .../u2++_conformer_wenet/pytorch/README.md    |   6 +
 .../unified_conformer_wenet/pytorch/README.md |   6 +
 .../fastspeech2/paddlepaddle/README.md        |   6 +
 .../hifigan/paddlepaddle/README.md            |   6 +
 .../tacotron2/pytorch/README.md               |   6 +
 .../speech_synthesis/vqmivc/pytorch/README.md |   6 +
 .../waveglow/pytorch/README.md                |   6 +
 .../hashnerf/pytorch/README.md                |   6 +
 cv/3d_detection/bevformer/pytorch/README.md   |   6 +
 cv/3d_detection/centerpoint/pytorch/README.md |   6 +
 cv/3d_detection/paconv/pytorch/README.md      |   6 +
 .../part_a2_anchor/pytorch/README.md          |   6 +
 .../part_a2_free/pytorch/README.md            |   6 +
 cv/3d_detection/pointnet2/pytorch/README.md   |   6 +
 .../pointpillars/pytorch/README.md            |   6 +
 cv/3d_detection/pointrcnn/pytorch/README.md   |   6 +
 .../pointrcnn_iou/pytorch/README.md           | 156 ++++++++---------
 cv/3d_detection/second/pytorch/README.md      | 156 ++++++++---------
 cv/3d_detection/second_iou/pytorch/README.md  | 158 +++++++++---------
 cv/distiller/cwd/pytorch/README.md            |   6 +
 cv/distiller/rkd/pytorch/README.md            |   6 +
 cv/distiller/wsld/pytorch/README.md           |   6 +
 .../retinaface/pytorch/README.md              |   6 +
 cv/face_recognition/arcface/pytorch/README.md |   6 +
 .../blazeface/paddlepaddle/README.md          |   6 +
 cv/face_recognition/cosface/pytorch/README.md |   6 +
 cv/face_recognition/facenet/pytorch/README.md |   6 +
 .../facenet/tensorflow/README.md              |   6 +
 cv/gnn/gat/paddlepaddle/README.md             |   6 +
 cv/gnn/gcn/mindspore/README.md                |   6 +
 cv/gnn/gcn/paddlepaddle/README.md             |   6 +
 cv/gnn/graphsage/paddlepaddle/README.md       |   6 +
 cv/image_generation/dcgan/mindspore/README.md |   6 +
 .../pix2pix/paddlepaddle/README.md            |   6 +
 .../solo/pytorch/README.md                    |   6 +
 .../solov2/pytorch/README.md                  |   6 +
 .../yolact/pytorch/README.md                  |   6 +
 .../bytetrack/paddlepaddle/README.md          |   6 +
 .../deep_sort/pytorch/README.md               |   6 +
 .../fairmot/pytorch/README.md                 |   6 +
 cv/ocr/crnn/mindspore/README.md               |   6 +
 cv/ocr/crnn/paddlepaddle/README.md            |   6 +
 cv/ocr/dbnet/pytorch/README.md                |   6 +
 cv/ocr/dbnetpp/paddlepaddle/README.md         |   6 +
 cv/ocr/dbnetpp/pytorch/README.md              |   6 +
 cv/ocr/pp-ocr-db/paddlepaddle/README.md       |   6 +
 cv/ocr/pp-ocr-east/paddlepaddle/README.md     |   6 +
 cv/ocr/pse/paddlepaddle/README.md             |   6 +
 cv/ocr/sar/pytorch/README.md                  |   6 +
 cv/ocr/sast/paddlepaddle/README.md            |   6 +
 cv/ocr/satrn/pytorch/base/README.md           |   6 +
 cv/point_cloud/point-bert/pytorch/README.md   |   6 +
 cv/pose/alphapose/pytorch/README.md           |   6 +
 cv/pose/hrnet/paddlepaddle/README.md          |   6 +
 cv/pose/hrnet/pytorch/README.md               |   6 +
 cv/pose/openpose/mindspore/README.md          |   6 +
 .../mae/pytorch/README.md                     |   6 +
 .../apcnet/pytorch/README.md                  |   6 +
 .../att_unet/pytorch/README.md                |   6 +
 .../bisenet/pytorch/README.md                 |   6 +
 .../bisenetv2/paddlepaddle/README.md          |   6 +
 .../bisenetv2/pytorch/README.md               |   6 +
 .../cgnet/pytorch/README.md                   |   6 +
 .../contextnet/pytorch/README.md              |   6 +
 .../dabnet/pytorch/README.md                  |   6 +
 .../danet/pytorch/README.md                   |   6 +
 .../ddrnet/pytorch/README.md                  |   6 +
 .../deeplabv3/mindspore/README.md             |   6 +
 .../deeplabv3/paddlepaddle/README.md          |   6 +
 .../deeplabv3/pytorch/README.md               |   6 +
 .../deeplabv3plus/paddlepaddle/README.md      |   6 +
 .../deeplabv3plus/tensorflow/README.md        |   6 +
 .../denseaspp/pytorch/README.md               |   6 +
 .../dfanet/pytorch/README.md                  |   6 +
 .../dnlnet/paddlepaddle/README.md             |   6 +
 .../dunet/pytorch/README.md                   |   6 +
 .../encnet/pytorch/README.md                  |   6 +
 .../enet/pytorch/README.md                    |   6 +
 .../erfnet/pytorch/README.md                  |   6 +
 .../espnet/pytorch/README.md                  |   6 +
 .../fastfcn/paddlepaddle/README.md            |   6 +
 .../fastscnn/pytorch/README.md                |   6 +
 .../fcn/pytorch/README.md                     |   6 +
 .../fpenet/pytorch/README.md                  |   6 +
 .../gcnet/pytorch/README.md                   |   6 +
 .../hardnet/pytorch/README.md                 |   6 +
 .../icnet/pytorch/README.md                   |   6 +
 .../lednet/pytorch/README.md                  |   6 +
 .../linknet/pytorch/README.md                 |   6 +
 .../mask2former/pytorch/README.md             |   6 +
 .../mobileseg/paddlepaddle/README.md          |   6 +
 .../ocnet/pytorch/README.md                   |   6 +
 .../ocrnet/paddlepaddle/README.md             |   6 +
 .../ocrnet/pytorch/README.md                  |   6 +
 .../pp_humansegv1/paddlepaddle/README.md      |   6 +
 .../pp_humansegv2/paddlepaddle/README.md      |   6 +
 .../pp_liteseg/paddlepaddle/README.md         |   6 +
 .../psanet/pytorch/README.md                  |   6 +
 .../pspnet/pytorch/README.md                  |   6 +
 .../refinenet/pytorch/README.md               |   6 +
 .../segnet/pytorch/README.md                  |   6 +
 .../stdc/paddlepaddle/README.md               |   6 +
 .../stdc/pytorch/README.md                    |   6 +
 .../unet++/pytorch/README.md                  |   6 +
 .../unet/paddlepaddle/README.md               |   6 +
 .../unet/pytorch/README.md                    |   6 +
 .../unet3d/pytorch/README.md                  |   6 +
 .../vnet/tensorflow/README.md                 |   6 +
 .../basicvsr++/pytorch/README.md              |   6 +
 .../basicvsr/pytorch/README.md                |   6 +
 cv/super_resolution/esrgan/pytorch/README.md  | 108 ++++++------
 cv/super_resolution/liif/pytorch/README.md    |   6 +
 .../real_basicvsr/pytorch/README.md           |   6 +
 cv/super_resolution/ttsr/pytorch/README.md    |   6 +
 cv/super_resolution/ttvsr/pytorch/README.md   |   6 +
 docs/MODEL_TEMPLATE.md                        |   8 +-
 .../water_se_e2_a/tensorflow/README.md        |  28 +++-
 .../clip/pytorch/README.md                    |   6 +
 .../controlnet/pytorch/README.md              |   6 +
 .../diffusion_model/ddpm/pytorch/README.md    |   6 +
 .../stable-diffusion-1.4/pytorch/README.md    |   6 +
 .../stable-diffusion-1.5/pytorch/README.md    |   6 +
 .../stable-diffusion-2.1/pytorch/README.md    |   6 +
 .../stable-diffusion-3/pytorch/README.md      |   6 +
 .../stable-diffusion-xl/pytorch/README.md     |   6 +
 .../blip/pytorch/README.md                    |   6 +
 .../l-verse/pytorch/README.md                 |   6 +
 .../llava-1.5/pytorch/README.md               |   6 +
 .../moe-llava-phi2-2.7b/pytorch/README.md     |   6 +
 .../moe-llava-qwen-1.8b/pytorch/README.md     |   6 +
 .../moe-llava-stablelm-1.6b/pytorch/README.md |   6 +
 .../{GLMForMultiTokenCloze => }/README.md     |   6 +
 nlp/dialogue_generation/cpm/pytorch/README.md |   6 +
 .../bart_fairseq/pytorch/README.md            |   6 +
 nlp/language_model/bert/mindspore/README.md   |   6 +
 .../bert/paddlepaddle/README.md               |   6 +
 nlp/language_model/bert/pytorch/README.md     |   6 +
 .../bert/tensorflow/{base => }/README.md      | 142 ++++++++--------
 .../roberta_fairseq/pytorch/README.md         |   6 +
 .../xlnet/paddlepaddle/README.md              |   6 +
 nlp/ner/bert/pytorch/README.md                |   6 +
 nlp/question_answering/bert/pytorch/README.md |   6 +
 .../bert/pytorch/README.md                    |   6 +
 .../ernie/paddlepaddle/README.md              |   6 +
 nlp/text_summarisation/bert/pytorch/README.md |   6 +
 .../convolutional_fairseq/pytorch/README.md   | 144 ++++++++--------
 nlp/translation/t5/pytorch/README.md          |   6 +
 .../transformer/paddlepaddle/README.md        |   6 +
 .../transformer_fairseq/pytorch/README.md     | 156 ++++++++---------
 .../graph_wavenet/pytorch/README.md           |   6 +
 .../kan/pytorch/README.md                     |   6 +
 .../network-slimming/pytorch/README.md        |   6 +
 .../deepfm/paddlepaddle/README.md             |   6 +
 .../dlrm/paddlepaddle/README.md               |   6 +
 .../dlrm/pytorch/README.md                    |   6 +
 .../ffm/paddlepaddle/README.md                |   6 +
 .../ncf/pytorch/README.md                     |   6 +
 .../wide_deep/paddlepaddle/README.md          |   6 +
 .../xdeepfm/paddlepaddle/README.md            |   6 +
 .../dqn/paddlepaddle/README.md                |   6 +
 165 files changed, 1490 insertions(+), 502 deletions(-)
 rename nlp/cloze_test/glm/pytorch/{GLMForMultiTokenCloze => }/README.md (85%)
 rename nlp/language_model/bert/tensorflow/{base => }/README.md (92%)

diff --git a/audio/speech_recognition/conformer/paddlepaddle/README.md b/audio/speech_recognition/conformer/paddlepaddle/README.md
index 05356b0f8..adebd5755 100644
--- a/audio/speech_recognition/conformer/paddlepaddle/README.md
+++ b/audio/speech_recognition/conformer/paddlepaddle/README.md
@@ -12,6 +12,12 @@ CNN based models achieving state-of-the-art accuracies. On the widely used Libri
 of 2.1%/4.3% without using a language model and 1.9%/3.9% with an external language model on test/testother. We also
 observe competitive performance of 2.7%/6.3% with a small model of only 10M parameters.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_recognition/conformer_wenet/pytorch/README.md b/audio/speech_recognition/conformer_wenet/pytorch/README.md
index 6d39b05d4..04a07b4b5 100755
--- a/audio/speech_recognition/conformer_wenet/pytorch/README.md
+++ b/audio/speech_recognition/conformer_wenet/pytorch/README.md
@@ -7,6 +7,12 @@ convolutional neural networks (CNNs) and transformers. It employs CNNs for local
 capture long-range dependencies in data. This combination allows the Conformer to efficiently handle both local patterns
 and global relationships, making it particularly effective for audio and speech tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/audio/speech_recognition/efficient_conformer_v2_wenet/pytorch/README.md b/audio/speech_recognition/efficient_conformer_v2_wenet/pytorch/README.md
index 1ac4b0f62..a159c2eb5 100644
--- a/audio/speech_recognition/efficient_conformer_v2_wenet/pytorch/README.md
+++ b/audio/speech_recognition/efficient_conformer_v2_wenet/pytorch/README.md
@@ -7,6 +7,12 @@ offering transformers a series of designs and optimizations for mobile accelerat
 The number of parameters and latency of the model are critical for resource-constrained hardware,
 so EfficientFormerV2 combines a fine-grained joint search strategy to propose an efficient network with low latency and size.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/audio/speech_recognition/rnnt/pytorch/README.md b/audio/speech_recognition/rnnt/pytorch/README.md
index 17dff16a0..c0352bf7e 100644
--- a/audio/speech_recognition/rnnt/pytorch/README.md
+++ b/audio/speech_recognition/rnnt/pytorch/README.md
@@ -10,6 +10,12 @@ combines these representations. RNN-T handles variable-length input/output seque
 during training. It's particularly effective for speech recognition as it can process continuous audio streams and
 output text in real-time, achieving state-of-the-art performance on various benchmarks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_recognition/transformer_wenet/pytorch/README.md b/audio/speech_recognition/transformer_wenet/pytorch/README.md
index 3138a0244..b9e8f2090 100755
--- a/audio/speech_recognition/transformer_wenet/pytorch/README.md
+++ b/audio/speech_recognition/transformer_wenet/pytorch/README.md
@@ -8,6 +8,12 @@ parallel (as opposed to sequentially) and capture complex dependencies in data,
 sequence. Transformers have since become the foundation for state-of-the-art models in various tasks, especially in
 natural language processing, such as the BERT and GPT series.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/audio/speech_recognition/u2++_conformer_wenet/pytorch/README.md b/audio/speech_recognition/u2++_conformer_wenet/pytorch/README.md
index 555bc945f..ef640d191 100755
--- a/audio/speech_recognition/u2++_conformer_wenet/pytorch/README.md
+++ b/audio/speech_recognition/u2++_conformer_wenet/pytorch/README.md
@@ -6,6 +6,12 @@ U2++, an enhanced version of U2 to further improve the accuracy. The core idea o
 backward information of the labeling sequences at the same time at training to learn richer information, and combine the
 forward and backward prediction at decoding to give more accurate recognition results.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/audio/speech_recognition/unified_conformer_wenet/pytorch/README.md b/audio/speech_recognition/unified_conformer_wenet/pytorch/README.md
index 6a2eb85f0..420fa9f8e 100755
--- a/audio/speech_recognition/unified_conformer_wenet/pytorch/README.md
+++ b/audio/speech_recognition/unified_conformer_wenet/pytorch/README.md
@@ -7,6 +7,12 @@ Unified Conformer is an architecture that has become state-of-the-art in the fie
 Processing and Computer Vision tasks thanks to its powerful self-attention mechanism¹. The Conformer architecture has
 been modified from ASR to Automatic Speaker Verification (ASV) with very minor changes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/audio/speech_synthesis/fastspeech2/paddlepaddle/README.md b/audio/speech_synthesis/fastspeech2/paddlepaddle/README.md
index bce73c30e..eaf897aca 100644
--- a/audio/speech_synthesis/fastspeech2/paddlepaddle/README.md
+++ b/audio/speech_synthesis/fastspeech2/paddlepaddle/README.md
@@ -8,6 +8,12 @@ waveform from text in parallel, enjoying the benefit of fully end-to-end inferen
 FastSpeech 2 achieves a 3x training speed-up over FastSpeech, and FastSpeech 2s enjoys even faster inference speed; 2)
 FastSpeech 2 and 2s outperform FastSpeech in voice quality, and FastSpeech 2 can even surpass autoregressive models.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_synthesis/hifigan/paddlepaddle/README.md b/audio/speech_synthesis/hifigan/paddlepaddle/README.md
index d9e0063f4..525f154c7 100644
--- a/audio/speech_synthesis/hifigan/paddlepaddle/README.md
+++ b/audio/speech_synthesis/hifigan/paddlepaddle/README.md
@@ -6,6 +6,12 @@ HiFiGAN is a commonly used vocoder in academia and industry in recent years, whi
 generated by acoustic models into high-quality audio. This vocoder uses generative adversarial networks as the basis for
 generating models.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_synthesis/tacotron2/pytorch/README.md b/audio/speech_synthesis/tacotron2/pytorch/README.md
index b5bc8ef8f..e39cc657c 100644
--- a/audio/speech_synthesis/tacotron2/pytorch/README.md
+++ b/audio/speech_synthesis/tacotron2/pytorch/README.md
@@ -9,6 +9,12 @@ vocoder to produce high-quality audio. The model achieves near-human speech qual
 learned acoustic representations, enabling more natural prosody and articulation while maintaining computational
 efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_synthesis/vqmivc/pytorch/README.md b/audio/speech_synthesis/vqmivc/pytorch/README.md
index 4afac17dc..10c70335d 100644
--- a/audio/speech_synthesis/vqmivc/pytorch/README.md
+++ b/audio/speech_synthesis/vqmivc/pytorch/README.md
@@ -9,6 +9,12 @@ inter-dependencies between speech components, VQMIVC achieves superior naturalne
 traditional methods. This unsupervised approach is particularly effective for retaining source linguistic content while
 accurately capturing target speaker characteristics.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/audio/speech_synthesis/waveglow/pytorch/README.md b/audio/speech_synthesis/waveglow/pytorch/README.md
index 00058b66e..8db58ca88 100644
--- a/audio/speech_synthesis/waveglow/pytorch/README.md
+++ b/audio/speech_synthesis/waveglow/pytorch/README.md
@@ -8,6 +8,12 @@ audio synthesis, without the need for auto-regression. WaveGlow is implemented u
 using only a single cost function: maximizing the likelihood of the training data, which makes the training procedure
 simple and stable.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d-reconstruction/hashnerf/pytorch/README.md b/cv/3d-reconstruction/hashnerf/pytorch/README.md
index 3b9044668..90ccd5de9 100644
--- a/cv/3d-reconstruction/hashnerf/pytorch/README.md
+++ b/cv/3d-reconstruction/hashnerf/pytorch/README.md
@@ -8,6 +8,12 @@ efficiency. Based on instant-ngp's approach, HashNeRF employs a grid encoder and
 high-quality rendering results. The model supports various datasets and custom scenes, making it suitable for
 applications in computer graphics, virtual reality, and 3D reconstruction tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/bevformer/pytorch/README.md b/cv/3d_detection/bevformer/pytorch/README.md
index 7783512dd..b08399ed6 100755
--- a/cv/3d_detection/bevformer/pytorch/README.md
+++ b/cv/3d_detection/bevformer/pytorch/README.md
@@ -9,6 +9,12 @@ approach achieves state-of-the-art performance on nuScenes dataset, matching LiD
 multiple perception tasks simultaneously, making it a versatile solution for comprehensive scene understanding in
 autonomous driving applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/centerpoint/pytorch/README.md b/cv/3d_detection/centerpoint/pytorch/README.md
index 27a9d5d67..ae2cc6a34 100644
--- a/cv/3d_detection/centerpoint/pytorch/README.md
+++ b/cv/3d_detection/centerpoint/pytorch/README.md
@@ -8,6 +8,12 @@ size, orientation, and velocity. A second stage refines these estimates using ad
 simplifies 3D tracking to greedy closest-point matching, achieving top performance on nuScenes and Waymo datasets while
 maintaining efficiency and simplicity in implementation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/paconv/pytorch/README.md b/cv/3d_detection/paconv/pytorch/README.md
index 69a10d090..8987c3222 100644
--- a/cv/3d_detection/paconv/pytorch/README.md
+++ b/cv/3d_detection/paconv/pytorch/README.md
@@ -8,6 +8,12 @@ Bank, with coefficients learned from point positions through ScoreNet. This data
 handle irregular point cloud data efficiently. PAConv integrates seamlessly with existing MLP-based pipelines, achieving
 state-of-the-art performance in classification and segmentation tasks while maintaining computational efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/part_a2_anchor/pytorch/README.md b/cv/3d_detection/part_a2_anchor/pytorch/README.md
index 3e45faedc..2a2f106f2 100644
--- a/cv/3d_detection/part_a2_anchor/pytorch/README.md
+++ b/cv/3d_detection/part_a2_anchor/pytorch/README.md
@@ -8,6 +8,12 @@ part locations using free part supervisions; second, it aggregates these parts t
 approach effectively captures object geometry, achieving state-of-the-art performance on the KITTI dataset while
 maintaining computational efficiency for practical applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/part_a2_free/pytorch/README.md b/cv/3d_detection/part_a2_free/pytorch/README.md
index 189ba0b7e..8cafdce0a 100644
--- a/cv/3d_detection/part_a2_free/pytorch/README.md
+++ b/cv/3d_detection/part_a2_free/pytorch/README.md
@@ -8,6 +8,12 @@ supervisions, then aggregating these parts to refine box scores and locations. T
 geometry through a novel RoI-aware point cloud pooling module, achieving state-of-the-art performance on the KITTI
 dataset while maintaining computational efficiency for practical applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/pointnet2/pytorch/README.md b/cv/3d_detection/pointnet2/pytorch/README.md
index bd3547177..5c351c022 100644
--- a/cv/3d_detection/pointnet2/pytorch/README.md
+++ b/cv/3d_detection/pointnet2/pytorch/README.md
@@ -8,6 +8,12 @@ multiple scales. The network adapts to varying point densities through novel set
 on complex scenes. PointNet++ excels in tasks like 3D object classification and segmentation by effectively capturing
 fine-grained geometric patterns in point clouds.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/pointpillars/pytorch/README.md b/cv/3d_detection/pointpillars/pytorch/README.md
index ce5c69321..4d1533aa9 100755
--- a/cv/3d_detection/pointpillars/pytorch/README.md
+++ b/cv/3d_detection/pointpillars/pytorch/README.md
@@ -8,6 +8,12 @@ networks for processing. This approach balances accuracy and speed, making it su
 autonomous driving. PointPillars achieves state-of-the-art performance on the KITTI dataset while maintaining
 computational efficiency through its pillar-based encoding and simplified network architecture.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/pointrcnn/pytorch/README.md b/cv/3d_detection/pointrcnn/pytorch/README.md
index 42d36c3dd..11bfb3a8b 100644
--- a/cv/3d_detection/pointrcnn/pytorch/README.md
+++ b/cv/3d_detection/pointrcnn/pytorch/README.md
@@ -8,6 +8,12 @@ it generates accurate 3D box proposals in a bottom-up manner. The second stage r
 achieves state-of-the-art performance on the KITTI dataset, demonstrating superior accuracy in 3D object detection
 tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/3d_detection/pointrcnn_iou/pytorch/README.md b/cv/3d_detection/pointrcnn_iou/pytorch/README.md
index fe3b2e578..20005746b 100644
--- a/cv/3d_detection/pointrcnn_iou/pytorch/README.md
+++ b/cv/3d_detection/pointrcnn_iou/pytorch/README.md
@@ -1,75 +1,81 @@
-# PointRCNN-IoU
-
-## Model Description
-
-PointRCNN-IoU is an enhanced version of the PointRCNN framework that incorporates Intersection over Union (IoU)
-optimization for 3D object detection. It processes raw point cloud data in two stages: first generating 3D proposals,
-then refining them with IoU-aware regression. This approach improves bounding box accuracy by directly optimizing the
-overlap between predicted and ground truth boxes. PointRCNN-IoU maintains the efficiency of its predecessor while
-achieving higher precision in 3D object detection tasks.
-
-## Model Preparation
-
-### Prepare Resources
-
-Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
-
-Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
-
-```bash
-OpenPCDet
-├── data
-│   ├── kitti
-│   │   │── ImageSets
-│   │   │── training
-│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
-│   │   │── testing
-│   │   │   ├──calib & velodyne & image_2
-├── pcdet
-├── tools
-```
-
-```bash
-# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
-cd <deepsparkhub_root>/toolbox/openpcdet
-python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
-```
-
-### Install Dependencies
-
-```bash
-## install libGL and libboost
-yum install mesa-libGL
-yum install boost-devel
-
-# Install numba
-pushd <deepsparkhub_root>/toolbox/numba
-python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
-pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
-popd
-
-# Install spconv
-pushd <deepsparkhub_root>/toolbox/spconv
-bash clean_spconv.sh
-bash build_spconv.sh
-bash install_spconv.sh
-popd
-
-# Install openpcdet
-pushd <deepsparkhub_root>/toolbox/openpcdet
-pip3 install -r requirements.txt
-bash build_openpcdet.sh
-bash install_openpcdet.sh
-popd
-```
-
-## Model Training
-
-```bash
-# Single GPU training
-cd tools/
-python3 train.py --cfg_file cfgs/kitti_models/pointrcnn_iou.yaml
-
-# Multiple GPU training
-bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/pointrcnn_iou.yaml
-```
+# PointRCNN-IoU
+
+## Model Description
+
+PointRCNN-IoU is an enhanced version of the PointRCNN framework that incorporates Intersection over Union (IoU)
+optimization for 3D object detection. It processes raw point cloud data in two stages: first generating 3D proposals,
+then refining them with IoU-aware regression. This approach improves bounding box accuracy by directly optimizing the
+overlap between predicted and ground truth boxes. PointRCNN-IoU maintains the efficiency of its predecessor while
+achieving higher precision in 3D object detection tasks.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.06  |
+
+## Model Preparation
+
+### Prepare Resources
+
+Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
+
+Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
+
+```bash
+OpenPCDet
+├── data
+│   ├── kitti
+│   │   │── ImageSets
+│   │   │── training
+│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
+│   │   │── testing
+│   │   │   ├──calib & velodyne & image_2
+├── pcdet
+├── tools
+```
+
+```bash
+# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
+cd <deepsparkhub_root>/toolbox/openpcdet
+python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
+```
+
+### Install Dependencies
+
+```bash
+## install libGL and libboost
+yum install mesa-libGL
+yum install boost-devel
+
+# Install numba
+pushd <deepsparkhub_root>/toolbox/numba
+python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
+pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
+popd
+
+# Install spconv
+pushd <deepsparkhub_root>/toolbox/spconv
+bash clean_spconv.sh
+bash build_spconv.sh
+bash install_spconv.sh
+popd
+
+# Install openpcdet
+pushd <deepsparkhub_root>/toolbox/openpcdet
+pip3 install -r requirements.txt
+bash build_openpcdet.sh
+bash install_openpcdet.sh
+popd
+```
+
+## Model Training
+
+```bash
+# Single GPU training
+cd tools/
+python3 train.py --cfg_file cfgs/kitti_models/pointrcnn_iou.yaml
+
+# Multiple GPU training
+bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/pointrcnn_iou.yaml
+```
diff --git a/cv/3d_detection/second/pytorch/README.md b/cv/3d_detection/second/pytorch/README.md
index e28fe30e9..b41d98ade 100644
--- a/cv/3d_detection/second/pytorch/README.md
+++ b/cv/3d_detection/second/pytorch/README.md
@@ -1,75 +1,81 @@
-# SECOND
-
-## Model Description
-
-SECOND is an efficient 3D object detection framework for LiDAR point cloud data, utilizing sparse convolutional networks
-to enhance information retention. It introduces improved sparse convolution methods for faster training and inference,
-along with novel angle loss regression for better orientation estimation. The framework also incorporates a unique data
-augmentation approach to boost convergence speed and performance. SECOND achieves state-of-the-art results on the KITTI
-benchmark while maintaining rapid inference, making it suitable for real-time applications like autonomous driving.
-
-## Model Preparation
-
-### Prepare Resources
-
-Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
-
-Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
-
-```bash
-OpenPCDet
-├── data
-│   ├── kitti
-│   │   │── ImageSets
-│   │   │── training
-│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
-│   │   │── testing
-│   │   │   ├──calib & velodyne & image_2
-├── pcdet
-├── tools
-```
-
-```bash
-# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
-cd <deepsparkhub_root>/toolbox/openpcdet
-python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
-```
-
-### Install Dependencies
-
-```bash
-## install libGL and libboost
-yum install mesa-libGL
-yum install boost-devel
-
-# Install numba
-pushd <deepsparkhub_root>/toolbox/numba
-python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
-pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
-popd
-
-# Install spconv
-pushd <deepsparkhub_root>/toolbox/spconv
-bash clean_spconv.sh
-bash build_spconv.sh
-bash install_spconv.sh
-popd
-
-# Install openpcdet
-pushd <deepsparkhub_root>/toolbox/openpcdet
-pip3 install -r requirements.txt
-bash build_openpcdet.sh
-bash install_openpcdet.sh
-popd
-```
-
-## Model Training
-
-```bash
-# Single GPU training
-cd tools/
-python3 train.py --cfg_file cfgs/kitti_models/second.yaml
-
-# Multiple GPU training
-bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/second.yaml
-```
+# SECOND
+
+## Model Description
+
+SECOND is an efficient 3D object detection framework for LiDAR point cloud data, utilizing sparse convolutional networks
+to enhance information retention. It introduces improved sparse convolution methods for faster training and inference,
+along with novel angle loss regression for better orientation estimation. The framework also incorporates a unique data
+augmentation approach to boost convergence speed and performance. SECOND achieves state-of-the-art results on the KITTI
+benchmark while maintaining rapid inference, making it suitable for real-time applications like autonomous driving.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.06  |
+
+## Model Preparation
+
+### Prepare Resources
+
+Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
+
+Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
+
+```bash
+OpenPCDet
+├── data
+│   ├── kitti
+│   │   │── ImageSets
+│   │   │── training
+│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
+│   │   │── testing
+│   │   │   ├──calib & velodyne & image_2
+├── pcdet
+├── tools
+```
+
+```bash
+# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
+cd <deepsparkhub_root>/toolbox/openpcdet
+python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
+```
+
+### Install Dependencies
+
+```bash
+## install libGL and libboost
+yum install mesa-libGL
+yum install boost-devel
+
+# Install numba
+pushd <deepsparkhub_root>/toolbox/numba
+python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
+pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
+popd
+
+# Install spconv
+pushd <deepsparkhub_root>/toolbox/spconv
+bash clean_spconv.sh
+bash build_spconv.sh
+bash install_spconv.sh
+popd
+
+# Install openpcdet
+pushd <deepsparkhub_root>/toolbox/openpcdet
+pip3 install -r requirements.txt
+bash build_openpcdet.sh
+bash install_openpcdet.sh
+popd
+```
+
+## Model Training
+
+```bash
+# Single GPU training
+cd tools/
+python3 train.py --cfg_file cfgs/kitti_models/second.yaml
+
+# Multiple GPU training
+bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/second.yaml
+```
diff --git a/cv/3d_detection/second_iou/pytorch/README.md b/cv/3d_detection/second_iou/pytorch/README.md
index a36cabb15..766aa11ae 100644
--- a/cv/3d_detection/second_iou/pytorch/README.md
+++ b/cv/3d_detection/second_iou/pytorch/README.md
@@ -1,76 +1,82 @@
-# SECOND-IoU
-
-## Model Description
-
-SECOND-IoU is an enhanced version of the SECOND framework that incorporates Intersection over Union (IoU) optimization
-for 3D object detection from LiDAR point clouds. It leverages sparse convolutional networks to efficiently process 3D
-data while maintaining spatial information. The model introduces IoU-aware regression to improve bounding box accuracy
-and orientation estimation. SECOND-IoU achieves state-of-the-art performance on 3D detection benchmarks, offering faster
-inference speeds and better precision than traditional methods, making it suitable for real-time applications like
-autonomous driving.
-
-## Model Preparation
-
-### Prepare Resources
-
-Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
-
-Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
-
-```bash
-OpenPCDet
-├── data
-│   ├── kitti
-│   │   │── ImageSets
-│   │   │── training
-│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
-│   │   │── testing
-│   │   │   ├──calib & velodyne & image_2
-├── pcdet
-├── tools
-```
-
-```bash
-# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
-cd <deepsparkhub_root>/toolbox/openpcdet
-python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
-```
-
-### Install Dependencies
-
-```bash
-## install libGL and libboost
-yum install mesa-libGL
-yum install boost-devel
-
-# Install numba
-pushd <deepsparkhub_root>/toolbox/numba
-python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
-pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
-popd
-
-# Install spconv
-pushd <deepsparkhub_root>/toolbox/spconv
-bash clean_spconv.sh
-bash build_spconv.sh
-bash install_spconv.sh
-popd
-
-# Install openpcdet
-pushd <deepsparkhub_root>/toolbox/openpcdet
-pip3 install -r requirements.txt
-bash build_openpcdet.sh
-bash install_openpcdet.sh
-popd
-```
-
-## Model Training
-
-```bash
-# Single GPU training
-cd tools/
-python3 train.py --cfg_file cfgs/kitti_models/second_iou.yaml
-
-# Multiple GPU training
-bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/second_iou.yaml
-```
+# SECOND-IoU
+
+## Model Description
+
+SECOND-IoU is an enhanced version of the SECOND framework that incorporates Intersection over Union (IoU) optimization
+for 3D object detection from LiDAR point clouds. It leverages sparse convolutional networks to efficiently process 3D
+data while maintaining spatial information. The model introduces IoU-aware regression to improve bounding box accuracy
+and orientation estimation. SECOND-IoU achieves state-of-the-art performance on 3D detection benchmarks, offering faster
+inference speeds and better precision than traditional methods, making it suitable for real-time applications like
+autonomous driving.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.06  |
+
+## Model Preparation
+
+### Prepare Resources
+
+Download the kitti dataset from <http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d>
+
+Download the "planes" subdataset from <https://drive.google.com/file/d/1d5mq0RXRnvHPVeKx6Q612z0YRO1t2wAp/view?usp=sharing>
+
+```bash
+OpenPCDet
+├── data
+│   ├── kitti
+│   │   │── ImageSets
+│   │   │── training
+│   │   │   ├──calib & velodyne & label_2 & image_2 & (optional: planes) & (optional: depth_2)
+│   │   │── testing
+│   │   │   ├──calib & velodyne & image_2
+├── pcdet
+├── tools
+```
+
+```bash
+# Modify the `DATA_PATH` in the kitti_dataset.yaml to your own
+cd <deepsparkhub_root>/toolbox/openpcdet
+python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml
+```
+
+### Install Dependencies
+
+```bash
+## install libGL and libboost
+yum install mesa-libGL
+yum install boost-devel
+
+# Install numba
+pushd <deepsparkhub_root>/toolbox/numba
+python3 setup.py bdist_wheel -d build_pip 2>&1 | tee compile.log
+pip3 install build_pip/numba-0.56.4-cp310-cp310-linux_x86_64.whl
+popd
+
+# Install spconv
+pushd <deepsparkhub_root>/toolbox/spconv
+bash clean_spconv.sh
+bash build_spconv.sh
+bash install_spconv.sh
+popd
+
+# Install openpcdet
+pushd <deepsparkhub_root>/toolbox/openpcdet
+pip3 install -r requirements.txt
+bash build_openpcdet.sh
+bash install_openpcdet.sh
+popd
+```
+
+## Model Training
+
+```bash
+# Single GPU training
+cd tools/
+python3 train.py --cfg_file cfgs/kitti_models/second_iou.yaml
+
+# Multiple GPU training
+bash scripts/dist_train.sh 16 --cfg_file cfgs/kitti_models/second_iou.yaml
+```
diff --git a/cv/distiller/cwd/pytorch/README.md b/cv/distiller/cwd/pytorch/README.md
index db68c3944..20fee28a4 100644
--- a/cv/distiller/cwd/pytorch/README.md
+++ b/cv/distiller/cwd/pytorch/README.md
@@ -8,6 +8,12 @@ student networks by transforming each channel's feature map into a probability m
 This approach focuses on the most salient regions of channel-wise maps, improving distillation efficiency and accuracy.
 CWD outperforms spatial distillation methods while requiring less computational cost during training.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/distiller/rkd/pytorch/README.md b/cv/distiller/rkd/pytorch/README.md
index 3bd0e331f..5887a8e4f 100755
--- a/cv/distiller/rkd/pytorch/README.md
+++ b/cv/distiller/rkd/pytorch/README.md
@@ -8,6 +8,12 @@ preserving the relationships (distance and angle) between embeddings. This appro
 learning tasks, where maintaining the relative structure of the embedding space is crucial. RKD enhances student model
 performance by capturing higher-order relational knowledge from the teacher.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/cv/distiller/wsld/pytorch/README.md b/cv/distiller/wsld/pytorch/README.md
index d3c2b3b26..a623d08e5 100644
--- a/cv/distiller/wsld/pytorch/README.md
+++ b/cv/distiller/wsld/pytorch/README.md
@@ -8,6 +8,12 @@ WSLD assigns different weights to each class based on their importance or diffic
 on challenging or critical classes. This approach improves the student model's performance, particularly in imbalanced
 datasets or tasks where certain classes require more attention.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_detection/retinaface/pytorch/README.md b/cv/face_detection/retinaface/pytorch/README.md
index e220ac6f4..320c60385 100644
--- a/cv/face_detection/retinaface/pytorch/README.md
+++ b/cv/face_detection/retinaface/pytorch/README.md
@@ -14,6 +14,12 @@ On the IJB-C test set, RetinaFace enables state of the art methods (ArcFace) to
 verification (TAR=89.59% for FAR=1e-6). (5) By employing light-weight backbone networks, RetinaFace can run real-time on
 a single CPU core for a VGA-resolution image.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_recognition/arcface/pytorch/README.md b/cv/face_recognition/arcface/pytorch/README.md
index d11fd055c..c0567028b 100644
--- a/cv/face_recognition/arcface/pytorch/README.md
+++ b/cv/face_recognition/arcface/pytorch/README.md
@@ -7,6 +7,12 @@ discriminative features for face recognition. The proposed ArcFace has a clear g
 correspondence to the geodesic distance on the hypersphere. ArcFace consistently outperforms the state-of-the-art and
 can be easily implemented with negligible computational overhead
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_recognition/blazeface/paddlepaddle/README.md b/cv/face_recognition/blazeface/paddlepaddle/README.md
index 60772cbb5..d7bf66467 100644
--- a/cv/face_recognition/blazeface/paddlepaddle/README.md
+++ b/cv/face_recognition/blazeface/paddlepaddle/README.md
@@ -5,6 +5,12 @@
 BlazeFace is Google Research published face detection model. It's lightweight but good performance, and tailored for
 mobile GPU inference. It runs at a speed of 200-1000+ FPS on flagship devices.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_recognition/cosface/pytorch/README.md b/cv/face_recognition/cosface/pytorch/README.md
index 3ac73556b..1c66b6182 100644
--- a/cv/face_recognition/cosface/pytorch/README.md
+++ b/cv/face_recognition/cosface/pytorch/README.md
@@ -6,6 +6,12 @@ CosFace is a face recognition model that achieves state-of-the-art results by in
 loss function when training the neural network, which learns highly discriminative facial embeddings by maximizing
 inter-class differences and minimizing intra-class variations.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_recognition/facenet/pytorch/README.md b/cv/face_recognition/facenet/pytorch/README.md
index 9d0d66112..d9ffe0005 100644
--- a/cv/face_recognition/facenet/pytorch/README.md
+++ b/cv/face_recognition/facenet/pytorch/README.md
@@ -8,6 +8,12 @@ closer together than those of different individuals. Facenet excels in tasks lik
 clustering, offering high accuracy and efficiency. Its compact embeddings make it scalable for large-scale applications
 in security and identity verification.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/face_recognition/facenet/tensorflow/README.md b/cv/face_recognition/facenet/tensorflow/README.md
index 16dee2400..09e84498e 100644
--- a/cv/face_recognition/facenet/tensorflow/README.md
+++ b/cv/face_recognition/facenet/tensorflow/README.md
@@ -8,6 +8,12 @@ closer together than those of different individuals. Facenet excels in tasks lik
 clustering, offering high accuracy and efficiency. Its compact embeddings make it scalable for large-scale applications
 in security and identity verification.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/gnn/gat/paddlepaddle/README.md b/cv/gnn/gat/paddlepaddle/README.md
index 6429faf2a..e000ccd6e 100644
--- a/cv/gnn/gat/paddlepaddle/README.md
+++ b/cv/gnn/gat/paddlepaddle/README.md
@@ -8,6 +8,12 @@ to neighboring nodes through attention coefficients, allowing for more flexible
 approach enables the model to handle varying neighborhood sizes and capture complex relationships in graph data, making
 it particularly effective for tasks like node classification and graph-based prediction problems.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/gnn/gcn/mindspore/README.md b/cv/gnn/gcn/mindspore/README.md
index 8e48864cd..403e86fce 100755
--- a/cv/gnn/gcn/mindspore/README.md
+++ b/cv/gnn/gcn/mindspore/README.md
@@ -7,6 +7,12 @@ data. A scalable approach based on an efficient variant of convolutional neural
 graphs was presented. The model scales linearly in the number of graph edges and learns hidden layer representations
 that encode both local graph structure and features of nodes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/gnn/gcn/paddlepaddle/README.md b/cv/gnn/gcn/paddlepaddle/README.md
index 63cb50c18..2eb4ee3d1 100644
--- a/cv/gnn/gcn/paddlepaddle/README.md
+++ b/cv/gnn/gcn/paddlepaddle/README.md
@@ -7,6 +7,12 @@ data. A scalable approach based on an efficient variant of convolutional neural
 graphs was presented. The model scales linearly in the number of graph edges and learns hidden layer representations
 that encode both local graph structure and features of nodes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/gnn/graphsage/paddlepaddle/README.md b/cv/gnn/graphsage/paddlepaddle/README.md
index fe7a86ba6..c8e2e25c8 100644
--- a/cv/gnn/graphsage/paddlepaddle/README.md
+++ b/cv/gnn/graphsage/paddlepaddle/README.md
@@ -8,6 +8,12 @@ sampling and aggregating features from a node's local neighborhood. This approac
 unseen nodes and graphs, making it particularly effective for dynamic graphs and large-scale applications like social
 network analysis and recommendation systems.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/image_generation/dcgan/mindspore/README.md b/cv/image_generation/dcgan/mindspore/README.md
index e5c2825da..7f365275e 100644
--- a/cv/image_generation/dcgan/mindspore/README.md
+++ b/cv/image_generation/dcgan/mindspore/README.md
@@ -5,6 +5,12 @@
 The deep convolutional generative adversarial networks (DCGANs) first introduced CNN into the GAN structure, and the
 strong feature extraction ability of convolution layer was used to improve the generation effect of GAN.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/image_generation/pix2pix/paddlepaddle/README.md b/cv/image_generation/pix2pix/paddlepaddle/README.md
index 684ed993a..84cfbc043 100755
--- a/cv/image_generation/pix2pix/paddlepaddle/README.md
+++ b/cv/image_generation/pix2pix/paddlepaddle/README.md
@@ -8,6 +8,12 @@ information to the generation network, Pix2pix uses another style of image as th
 generation network, so the fake image is related to another style of image which is input as supervision information,
 thus realizing the process of image translation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/instance_segmentation/solo/pytorch/README.md b/cv/instance_segmentation/solo/pytorch/README.md
index 237fd3b69..41a1ce85a 100644
--- a/cv/instance_segmentation/solo/pytorch/README.md
+++ b/cv/instance_segmentation/solo/pytorch/README.md
@@ -8,6 +8,12 @@ categories to each pixel based on an object's location and size. Unlike traditio
 instance masks without complex post-processing or region proposals. This approach achieves competitive accuracy with
 Mask R-CNN while offering a simpler and more flexible framework for instance-level recognition tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/instance_segmentation/solov2/pytorch/README.md b/cv/instance_segmentation/solov2/pytorch/README.md
index 01939217a..e6fbc7924 100644
--- a/cv/instance_segmentation/solov2/pytorch/README.md
+++ b/cv/instance_segmentation/solov2/pytorch/README.md
@@ -8,6 +8,12 @@ SOLOv2 introduces Matrix NMS, a faster non-maximum suppression technique that pr
 architecture achieves state-of-the-art performance in both speed and accuracy, with a lightweight version running at
 31.3 FPS. It serves as a strong baseline for various instance-level recognition tasks beyond segmentation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/instance_segmentation/yolact/pytorch/README.md b/cv/instance_segmentation/yolact/pytorch/README.md
index c06cee1a5..43d1d57e9 100644
--- a/cv/instance_segmentation/yolact/pytorch/README.md
+++ b/cv/instance_segmentation/yolact/pytorch/README.md
@@ -8,6 +8,12 @@ instance masks. This approach enables fast processing while maintaining competit
 performance with deformable convolutions and optimized prediction heads. The model achieves real-time speeds on single
 GPUs, making it suitable for applications requiring quick instance segmentation in video streams or interactive systems.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/multi_object_tracking/bytetrack/paddlepaddle/README.md b/cv/multi_object_tracking/bytetrack/paddlepaddle/README.md
index 4e7ef4662..a49ce47f9 100644
--- a/cv/multi_object_tracking/bytetrack/paddlepaddle/README.md
+++ b/cv/multi_object_tracking/bytetrack/paddlepaddle/README.md
@@ -9,6 +9,12 @@ performance on benchmarks like MOT17, with high MOTA, IDF1, and HOTA scores whil
 speeds. Its simple yet effective design makes it a robust solution for various object tracking applications in video
 analysis.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/multi_object_tracking/deep_sort/pytorch/README.md b/cv/multi_object_tracking/deep_sort/pytorch/README.md
index bf4debfe1..2026a0859 100644
--- a/cv/multi_object_tracking/deep_sort/pytorch/README.md
+++ b/cv/multi_object_tracking/deep_sort/pytorch/README.md
@@ -8,6 +8,12 @@ especially in complex scenarios with occlusions. DeepSORT uses a Kalman filter f
 detections using both motion and appearance cues. This approach improves tracking consistency and reduces identity
 switches, making it particularly effective for person tracking in crowded scenes and video surveillance applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/multi_object_tracking/fairmot/pytorch/README.md b/cv/multi_object_tracking/fairmot/pytorch/README.md
index e4167dd92..b852dbd06 100644
--- a/cv/multi_object_tracking/fairmot/pytorch/README.md
+++ b/cv/multi_object_tracking/fairmot/pytorch/README.md
@@ -8,6 +8,12 @@ Operating on high-resolution feature maps, FairMOT achieves fairness between det
 improved tracking accuracy. Its joint learning approach eliminates the need for cascaded processing, making it more
 efficient and effective for complex tracking scenarios in crowded environments.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/crnn/mindspore/README.md b/cv/ocr/crnn/mindspore/README.md
index 649384425..8cbaaade3 100644
--- a/cv/ocr/crnn/mindspore/README.md
+++ b/cv/ocr/crnn/mindspore/README.md
@@ -9,6 +9,12 @@ without character segmentation or horizontal scaling, making it versatile for bo
 recognition tasks. Its compact architecture and unified framework make it practical for real-world applications like
 document analysis and OCR.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/crnn/paddlepaddle/README.md b/cv/ocr/crnn/paddlepaddle/README.md
index 6c9b8b167..b7e637216 100644
--- a/cv/ocr/crnn/paddlepaddle/README.md
+++ b/cv/ocr/crnn/paddlepaddle/README.md
@@ -9,6 +9,12 @@ without character segmentation or horizontal scaling, making it versatile for bo
 recognition tasks. Its compact architecture and unified framework make it practical for real-world applications like
 document analysis and OCR.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/dbnet/pytorch/README.md b/cv/ocr/dbnet/pytorch/README.md
index 8202dc43e..9ba1bd781 100755
--- a/cv/ocr/dbnet/pytorch/README.md
+++ b/cv/ocr/dbnet/pytorch/README.md
@@ -13,6 +13,12 @@ five benchmark datasets, which consistently achieves state-of-the-art results, i
 speed. In particular, with a light-weight backbone, the performance improvements by DB are significant so that we can
 look for an ideal tradeoff between detection accuracy and efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/dbnetpp/paddlepaddle/README.md b/cv/ocr/dbnetpp/paddlepaddle/README.md
index ad038cda1..1491fbba3 100644
--- a/cv/ocr/dbnetpp/paddlepaddle/README.md
+++ b/cv/ocr/dbnetpp/paddlepaddle/README.md
@@ -8,6 +8,12 @@ simplifying post-processing and improving accuracy. The ASF module enhances scal
 multi-scale features. This architecture enables DBNet++ to detect text of arbitrary shapes and extreme aspect ratios
 efficiently, achieving state-of-the-art performance in both accuracy and speed across various text detection benchmarks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/dbnetpp/pytorch/README.md b/cv/ocr/dbnetpp/pytorch/README.md
index ece02b7e8..c5cb0eca8 100644
--- a/cv/ocr/dbnetpp/pytorch/README.md
+++ b/cv/ocr/dbnetpp/pytorch/README.md
@@ -8,6 +8,12 @@ simplifying post-processing and improving accuracy. The ASF module enhances scal
 multi-scale features. This architecture enables DBNet++ to detect text of arbitrary shapes and extreme aspect ratios
 efficiently, achieving state-of-the-art performance in both accuracy and speed across various text detection benchmarks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/pp-ocr-db/paddlepaddle/README.md b/cv/ocr/pp-ocr-db/paddlepaddle/README.md
index 0d1b59ca9..765aefdba 100644
--- a/cv/ocr/pp-ocr-db/paddlepaddle/README.md
+++ b/cv/ocr/pp-ocr-db/paddlepaddle/README.md
@@ -8,6 +8,12 @@ model is optimized for real-time performance and can handle diverse text layouts
 particularly effective in document analysis and scene text recognition tasks, offering a balance between accuracy and
 computational efficiency for practical OCR applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/pp-ocr-east/paddlepaddle/README.md b/cv/ocr/pp-ocr-east/paddlepaddle/README.md
index d842f8308..371fee6de 100644
--- a/cv/ocr/pp-ocr-east/paddlepaddle/README.md
+++ b/cv/ocr/pp-ocr-east/paddlepaddle/README.md
@@ -8,6 +8,12 @@ scene images. The model is designed for real-time performance and can handle tex
 PP-OCR-EAST is particularly effective in complex scenarios, offering a balance between detection accuracy and
 computational efficiency for practical OCR applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/pse/paddlepaddle/README.md b/cv/ocr/pse/paddlepaddle/README.md
index 413c7f087..d4d4f5552 100644
--- a/cv/ocr/pse/paddlepaddle/README.md
+++ b/cv/ocr/pse/paddlepaddle/README.md
@@ -8,6 +8,12 @@ expansion algorithm. PSE effectively handles complex scenarios like curved text
 architecture combines feature pyramid networks with a novel post-processing method, making it particularly suitable for
 detecting text in diverse orientations and layouts with high accuracy.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/sar/pytorch/README.md b/cv/ocr/sar/pytorch/README.md
index cee369f63..e0e9fd312 100755
--- a/cv/ocr/sar/pytorch/README.md
+++ b/cv/ocr/sar/pytorch/README.md
@@ -10,6 +10,12 @@ off-the-shelf neural network components and only word-level annotations. It is c
 LSTM-based encoder-decoder framework and a 2-dimensional attention module. Despite its simplicity, the proposed method
 is robust and achieves state-of-the-art performance on both regular and irregular scene text recognition benchmarks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/sast/paddlepaddle/README.md b/cv/ocr/sast/paddlepaddle/README.md
index f98f954b7..8fb092421 100644
--- a/cv/ocr/sast/paddlepaddle/README.md
+++ b/cv/ocr/sast/paddlepaddle/README.md
@@ -11,6 +11,12 @@ be highly effective across several benchmarks like ICDAR2015 and SCUT-CTW1500, S
 also operates efficiently, achieving significant performance metrics such as running at 27.63 FPS on a NVIDIA Titan Xp
 with a high detection accuracy, making it a notable solution for arbitrary-shaped text detection challenges.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/ocr/satrn/pytorch/base/README.md b/cv/ocr/satrn/pytorch/base/README.md
index f3ce80cdd..34d69f3b7 100755
--- a/cv/ocr/satrn/pytorch/base/README.md
+++ b/cv/ocr/satrn/pytorch/base/README.md
@@ -9,6 +9,12 @@ enables it to handle complex text arrangements and large inter-character spacing
 outperforms traditional methods in recognizing irregular texts, making it valuable for real-world applications like sign
 and logo recognition.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/point_cloud/point-bert/pytorch/README.md b/cv/point_cloud/point-bert/pytorch/README.md
index 71231e2c1..0dc7a5c9b 100644
--- a/cv/point_cloud/point-bert/pytorch/README.md
+++ b/cv/point_cloud/point-bert/pytorch/README.md
@@ -9,6 +9,12 @@ AutoEncoder (dVAE) to generate discrete point tokens containing meaningful local
 some patches of input point clouds and feed them into the backbone Transformer. The pre-training objective is to recover
 the original point tokens at the masked locations under the supervision of point tokens obtained by the Tokenizer.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/pose/alphapose/pytorch/README.md b/cv/pose/alphapose/pytorch/README.md
index 20ebc61f8..a4dae3c4e 100755
--- a/cv/pose/alphapose/pytorch/README.md
+++ b/cv/pose/alphapose/pytorch/README.md
@@ -8,6 +8,12 @@ mAP) on COCO dataset and 80+ mAP (82.1 mAP) on MPII dataset. To match poses that
 frames, we also provide an efficient online pose tracker called Pose Flow. It is the first open-source online pose
 tracker that achieves both 60+ mAP (66.5 mAP) and 50+ MOTA (58.3 MOTA) on PoseTrack Challenge dataset.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/pose/hrnet/paddlepaddle/README.md b/cv/pose/hrnet/paddlepaddle/README.md
index 5bb76c392..edca0d096 100644
--- a/cv/pose/hrnet/paddlepaddle/README.md
+++ b/cv/pose/hrnet/paddlepaddle/README.md
@@ -9,6 +9,12 @@ one by one, and connect the multi-resolution streams in parallel. The resulting
 the nth stage contains n streams corresponding to n resolutions. The authors conduct repeated multi-resolution fusions
 by exchanging the information across the parallel streams over and over.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/pose/hrnet/pytorch/README.md b/cv/pose/hrnet/pytorch/README.md
index 02a1f0b93..ecd707118 100644
--- a/cv/pose/hrnet/pytorch/README.md
+++ b/cv/pose/hrnet/pytorch/README.md
@@ -9,6 +9,12 @@ one by one, and connect the multi-resolution streams in parallel. The resulting
 the nth stage contains n streams corresponding to n resolutions. The authors conduct repeated multi-resolution fusions
 by exchanging the information across the parallel streams over and over.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/pose/openpose/mindspore/README.md b/cv/pose/openpose/mindspore/README.md
index ed0616724..3cebd1c32 100644
--- a/cv/pose/openpose/mindspore/README.md
+++ b/cv/pose/openpose/mindspore/README.md
@@ -8,6 +8,12 @@ efficiency remains stable regardless of the number of people in an image. It sim
 associates them to individuals, making it particularly effective for scenarios with multiple people, such as crowd
 analysis and human-computer interaction applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/self_supervised_learning/mae/pytorch/README.md b/cv/self_supervised_learning/mae/pytorch/README.md
index 685f2d1b2..8963e7636 100644
--- a/cv/self_supervised_learning/mae/pytorch/README.md
+++ b/cv/self_supervised_learning/mae/pytorch/README.md
@@ -8,6 +8,12 @@ the encoder processes only the visible patches, and the lightweight decoder reco
 latent representation and mask tokens. MAE demonstrates that high masking ratios (e.g., 75%) can lead to robust feature
 learning, making it scalable and effective for various downstream vision tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/apcnet/pytorch/README.md b/cv/semantic_segmentation/apcnet/pytorch/README.md
index d139ec332..0138351af 100644
--- a/cv/semantic_segmentation/apcnet/pytorch/README.md
+++ b/cv/semantic_segmentation/apcnet/pytorch/README.md
@@ -7,6 +7,12 @@ representations with multiple well-designed Adaptive Context Modules (ACMs). Spe
 image representation as a guidance to estimate the local affinity coefficients for each sub-region. And then calculates
 a context vector with these affinities.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/att_unet/pytorch/README.md b/cv/semantic_segmentation/att_unet/pytorch/README.md
index 2469abaf1..5442e4c9f 100644
--- a/cv/semantic_segmentation/att_unet/pytorch/README.md
+++ b/cv/semantic_segmentation/att_unet/pytorch/README.md
@@ -9,6 +9,12 @@ modules, improving model sensitivity and accuracy. Attention U-Net efficiently p
 computational overhead, making it particularly effective for tasks requiring precise segmentation of complex anatomical
 structures.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/bisenet/pytorch/README.md b/cv/semantic_segmentation/bisenet/pytorch/README.md
index 0641e9fad..ae8489e29 100644
--- a/cv/semantic_segmentation/bisenet/pytorch/README.md
+++ b/cv/semantic_segmentation/bisenet/pytorch/README.md
@@ -7,6 +7,12 @@ spatial information and generate high-resolution features. Meanwhile, a Context
 is employed to obtain sufficient receptive field. On top of the two paths, we introduce a new Feature Fusion Module to
 combine features efficiently.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/bisenetv2/paddlepaddle/README.md b/cv/semantic_segmentation/bisenetv2/paddlepaddle/README.md
index aa8feb46c..f5f85d6ca 100644
--- a/cv/semantic_segmentation/bisenetv2/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/bisenetv2/paddlepaddle/README.md
@@ -10,6 +10,12 @@ supplied by the Detail Branch. Therefore, the Semantic Branch can be made very l
 fast-downsampling strategy. Both types of feature representation are merged to construct a stronger and more
 comprehensive feature representation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/bisenetv2/pytorch/README.md b/cv/semantic_segmentation/bisenetv2/pytorch/README.md
index 2b316072f..7e77b6844 100644
--- a/cv/semantic_segmentation/bisenetv2/pytorch/README.md
+++ b/cv/semantic_segmentation/bisenetv2/pytorch/README.md
@@ -10,6 +10,12 @@ supplied by the Detail Branch. Therefore, the Semantic Branch can be made very l
 fast-downsampling strategy. Both types of feature representation are merged to construct a stronger and more
 comprehensive feature representation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/cgnet/pytorch/README.md b/cv/semantic_segmentation/cgnet/pytorch/README.md
index afceb7958..b85fa82dd 100644
--- a/cv/semantic_segmentation/cgnet/pytorch/README.md
+++ b/cv/semantic_segmentation/cgnet/pytorch/README.md
@@ -6,6 +6,12 @@ A novel Context Guided Network (CGNet), which is a light-weight and efficient ne
 Under an equivalent number of parameters, the proposed CGNet significantly outperforms existing segmentation networks. 
 Specifically, without any post-processing and multi-scale testing, the proposed CGNet achieves 64.8% mean IoU on Cityscapes with less than 0.5 M parameters.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/contextnet/pytorch/README.md b/cv/semantic_segmentation/contextnet/pytorch/README.md
index 07f315f23..683be8dc7 100644
--- a/cv/semantic_segmentation/contextnet/pytorch/README.md
+++ b/cv/semantic_segmentation/contextnet/pytorch/README.md
@@ -7,6 +7,12 @@ pyramid representation to produce competitive semantic segmentation in real-time
 combines a deep network branch at low resolution that captures global context information efficiently with a shallow
 branch that focuses on high-resolution segmentation details.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/dabnet/pytorch/README.md b/cv/semantic_segmentation/dabnet/pytorch/README.md
index 40235ce2a..8dadbeb8f 100644
--- a/cv/semantic_segmentation/dabnet/pytorch/README.md
+++ b/cv/semantic_segmentation/dabnet/pytorch/README.md
@@ -6,6 +6,12 @@ A novel Depthwise Asymmetric Bottleneck (DAB) module, which efficiently adopts d
 Based on the DAB module, design a Depth-wise Asymmetric Bottleneck Network (DABNet) especially for real-time semantic segmentation.
 It creates sufficient receptive field and densely utilizes the contextual information.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/danet/pytorch/README.md b/cv/semantic_segmentation/danet/pytorch/README.md
index c0badffef..e00e50d7f 100644
--- a/cv/semantic_segmentation/danet/pytorch/README.md
+++ b/cv/semantic_segmentation/danet/pytorch/README.md
@@ -7,6 +7,12 @@ mechanism instead of simply stacking convolutions to compute the spatial attenti
 capture global information directly. DANet uses in parallel a position attention module and a channel attention module
 to capture feature dependencies in spatial and channel domains.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/ddrnet/pytorch/README.md b/cv/semantic_segmentation/ddrnet/pytorch/README.md
index cee8e72ab..0f89b4eaf 100644
--- a/cv/semantic_segmentation/ddrnet/pytorch/README.md
+++ b/cv/semantic_segmentation/ddrnet/pytorch/README.md
@@ -8,6 +8,12 @@ performed. Additionally, we design a new contextual information extractor named
 (DAPPM) to enlarge effective receptive fields and fuse multi-scale context based on low-resolution feature maps. Our
 method achieves a new state-of-the-art trade-off between accuracy and speed on both Cityscapes and CamVid dataset.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/deeplabv3/mindspore/README.md b/cv/semantic_segmentation/deeplabv3/mindspore/README.md
index c6c406a91..f8df4709f 100755
--- a/cv/semantic_segmentation/deeplabv3/mindspore/README.md
+++ b/cv/semantic_segmentation/deeplabv3/mindspore/README.md
@@ -7,6 +7,12 @@ keypoints of DeepLabV3: Its multi-grid atrous convolution makes it better to dea
 scales, and augmented ASPP makes image-level features available to capture long range information. This repository
 provides a script and recipe to DeepLabV3 model and achieve state-of-the-art performance.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/deeplabv3/paddlepaddle/README.md b/cv/semantic_segmentation/deeplabv3/paddlepaddle/README.md
index 6fa1239d1..d4bb64868 100644
--- a/cv/semantic_segmentation/deeplabv3/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/deeplabv3/paddlepaddle/README.md
@@ -6,6 +6,12 @@ DeepLabV3 is a semantic segmentation architecture that improves upon DeepLabV2 w
 problem of segmenting objects at multiple scales, modules are designed which employ atrous convolution in cascade or in
 parallel to capture multi-scale context by adopting multiple atrous rates.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/deeplabv3/pytorch/README.md b/cv/semantic_segmentation/deeplabv3/pytorch/README.md
index 0774e1920..890a17b15 100644
--- a/cv/semantic_segmentation/deeplabv3/pytorch/README.md
+++ b/cv/semantic_segmentation/deeplabv3/pytorch/README.md
@@ -6,6 +6,12 @@ DeepLabV3 is a semantic segmentation architecture that improves upon DeepLabV2 w
 problem of segmenting objects at multiple scales, modules are designed which employ atrous convolution in cascade or in
 parallel to capture multi-scale context by adopting multiple atrous rates. 
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/deeplabv3plus/paddlepaddle/README.md b/cv/semantic_segmentation/deeplabv3plus/paddlepaddle/README.md
index fe11612e9..a0b03d49d 100644
--- a/cv/semantic_segmentation/deeplabv3plus/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/deeplabv3plus/paddlepaddle/README.md
@@ -6,6 +6,12 @@ DeepLabv3 is a semantic segmentation architecture that improves upon DeepLabv2 w
 problem of segmenting objects at multiple scales, modules are designed which employ atrous convolution in cascade or in
 parallel to capture multi-scale context by adopting multiple atrous rates.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/deeplabv3plus/tensorflow/README.md b/cv/semantic_segmentation/deeplabv3plus/tensorflow/README.md
index afaa6bf65..2d76dfb19 100644
--- a/cv/semantic_segmentation/deeplabv3plus/tensorflow/README.md
+++ b/cv/semantic_segmentation/deeplabv3plus/tensorflow/README.md
@@ -7,6 +7,12 @@ encoder-decoder architecture. The network employs atrous convolution to capture
 effectively. It introduces a novel feature called the "ASPP" module, which utilizes parallel atrous convolutions to
 capture fine-grained details and global context simultaneously.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/denseaspp/pytorch/README.md b/cv/semantic_segmentation/denseaspp/pytorch/README.md
index a17fa3df1..219bfdc59 100644
--- a/cv/semantic_segmentation/denseaspp/pytorch/README.md
+++ b/cv/semantic_segmentation/denseaspp/pytorch/README.md
@@ -6,6 +6,12 @@ Densely connected Atrous Spatial Pyramid Pooling (DenseASPP), which connects a s
 dense way. Such that it generates multi-scale features that not only cover a larger scale range, but also cover that
 scale range densely, without significantly increasing the model size.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/dfanet/pytorch/README.md b/cv/semantic_segmentation/dfanet/pytorch/README.md
index c898bd2b7..b607aa74a 100644
--- a/cv/semantic_segmentation/dfanet/pytorch/README.md
+++ b/cv/semantic_segmentation/dfanet/pytorch/README.md
@@ -8,6 +8,12 @@ respectively. Based on the multi-scale feature propagation, DFANet substantially
 it still obtains sufficient receptive field and enhances the model learning ability, which strikes a balance between the
 speed and segmentation performance.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/dnlnet/paddlepaddle/README.md b/cv/semantic_segmentation/dnlnet/paddlepaddle/README.md
index 4722e6a23..a53cfbb72 100644
--- a/cv/semantic_segmentation/dnlnet/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/dnlnet/paddlepaddle/README.md
@@ -9,6 +9,12 @@ network's ability to capture contextual information while maintaining computatio
 superior performance in tasks requiring precise spatial understanding, such as urban scene segmentation, by effectively
 aggregating both local and global features.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/dunet/pytorch/README.md b/cv/semantic_segmentation/dunet/pytorch/README.md
index 42ec2f9cf..e47370fd8 100644
--- a/cv/semantic_segmentation/dunet/pytorch/README.md
+++ b/cv/semantic_segmentation/dunet/pytorch/README.md
@@ -8,6 +8,12 @@ designed to extract context information and enable precise localization by combi
 high-level ones. Furthermore, DUNet captures the retinal vessels at various shapes and scales by adaptively adjusting
 the receptive fields according to vessels' scales and shapes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/encnet/pytorch/README.md b/cv/semantic_segmentation/encnet/pytorch/README.md
index 3e59f3b55..08e58f822 100644
--- a/cv/semantic_segmentation/encnet/pytorch/README.md
+++ b/cv/semantic_segmentation/encnet/pytorch/README.md
@@ -5,6 +5,12 @@
 The Context Encoding Module, which captures the semantic context of scenes and selectively highlights class-dependent featuremaps. 
 The Context Encoding Module significantly improves semantic segmentation results with only marginal extra computation cost over FCN.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/enet/pytorch/README.md b/cv/semantic_segmentation/enet/pytorch/README.md
index 3ca986693..304caf27c 100644
--- a/cv/semantic_segmentation/enet/pytorch/README.md
+++ b/cv/semantic_segmentation/enet/pytorch/README.md
@@ -9,6 +9,12 @@ dilated convolutions, and spatial dropout. ENet's lightweight design makes it pa
 requiring fast inference on resource-constrained devices, such as mobile platforms or real-time video processing
 systems, without compromising segmentation quality.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/erfnet/pytorch/README.md b/cv/semantic_segmentation/erfnet/pytorch/README.md
index a8dc93345..56502973f 100644
--- a/cv/semantic_segmentation/erfnet/pytorch/README.md
+++ b/cv/semantic_segmentation/erfnet/pytorch/README.md
@@ -6,6 +6,12 @@ A deep architecture that is able to run in real-time while providing accurate se
 architecture is a novel layer that uses residual connections and factorized convolutions in order to remain efficient
 while retaining remarkable accuracy. 
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/espnet/pytorch/README.md b/cv/semantic_segmentation/espnet/pytorch/README.md
index d8b603aea..69d4525bf 100644
--- a/cv/semantic_segmentation/espnet/pytorch/README.md
+++ b/cv/semantic_segmentation/espnet/pytorch/README.md
@@ -6,6 +6,12 @@ ESPNet is a convolutional neural network for semantic segmentation of high resol
 ESPNet is based on a convolutional module, efficient spatial pyramid (ESP), which is efficient in terms of computation,
 memory, and power.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/fastfcn/paddlepaddle/README.md b/cv/semantic_segmentation/fastfcn/paddlepaddle/README.md
index 0e8b399ba..072b05030 100644
--- a/cv/semantic_segmentation/fastfcn/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/fastfcn/paddlepaddle/README.md
@@ -7,6 +7,12 @@ uses an efficient encoder-decoder architecture and depthwise separable convoluti
 simplified design allows FastFCN to run much faster than prior FCNs while maintaining good segmentation quality. FastFCN
 demonstrates real-time segmentation is possible with a carefully designed lightweight CNN architecture.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/fastscnn/pytorch/README.md b/cv/semantic_segmentation/fastscnn/pytorch/README.md
index b8b5c4a63..3ba25090b 100644
--- a/cv/semantic_segmentation/fastscnn/pytorch/README.md
+++ b/cv/semantic_segmentation/fastscnn/pytorch/README.md
@@ -7,6 +7,12 @@ resolution image data (1024x2048px) suited to efficient computation on embedded
 to downsample' module which computes low-level features for multiple resolution branches simultaneously. The network
 combines spatial detail at high resolution with deep features extracted at lower resolution.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/fcn/pytorch/README.md b/cv/semantic_segmentation/fcn/pytorch/README.md
index 4ae8295ca..fb145f400 100644
--- a/cv/semantic_segmentation/fcn/pytorch/README.md
+++ b/cv/semantic_segmentation/fcn/pytorch/README.md
@@ -9,6 +9,12 @@ connections are local. The network consists of a downsampling path, used to extr
 upsampling path, which allows for localization. FCNs also employ skip connections to recover the fine-grained spatial
 information lost in the downsampling path.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/fpenet/pytorch/README.md b/cv/semantic_segmentation/fpenet/pytorch/README.md
index b6c5c7795..f306b4ef3 100644
--- a/cv/semantic_segmentation/fpenet/pytorch/README.md
+++ b/cv/semantic_segmentation/fpenet/pytorch/README.md
@@ -7,6 +7,12 @@ Specifically, use a feature pyramid encoding block to encode multi-scale context
 convolutions in all stages of the encoder. A mutual embedding upsample module is introduced in the decoder to aggregate
 the high-level semantic features and low-level spatial details efficiently.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/gcnet/pytorch/README.md b/cv/semantic_segmentation/gcnet/pytorch/README.md
index 4e2ba8162..7f8016a71 100755
--- a/cv/semantic_segmentation/gcnet/pytorch/README.md
+++ b/cv/semantic_segmentation/gcnet/pytorch/README.md
@@ -6,6 +6,12 @@ A Global Context Network, or GCNet, utilises global context blocks to model long
 based on the Non-Local Network, but it modifies the architecture so less computation is required. Global context blocks
 are applied to multiple layers in a backbone network to construct the GCNet.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/hardnet/pytorch/README.md b/cv/semantic_segmentation/hardnet/pytorch/README.md
index 687683692..89dfb428b 100644
--- a/cv/semantic_segmentation/hardnet/pytorch/README.md
+++ b/cv/semantic_segmentation/hardnet/pytorch/README.md
@@ -6,6 +6,12 @@ The Harmonic Densely Connected Network to achieve high efficiency in terms of bo
 network achieves 35%, 36%, 30%, 32%, and 45% inference time reduction compared with FC-DenseNet-103, DenseNet-264,
 ResNet-50, ResNet-152, and SSD-VGG, respectively.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/icnet/pytorch/README.md b/cv/semantic_segmentation/icnet/pytorch/README.md
index fb48a24a6..271342021 100644
--- a/cv/semantic_segmentation/icnet/pytorch/README.md
+++ b/cv/semantic_segmentation/icnet/pytorch/README.md
@@ -6,6 +6,12 @@ An image cascade network (ICNet) that incorporates multi-resolution branches und
 in-depth analysis of the framework and introduce the cascade feature fusion unit to quickly achieve high-quality
 segmentation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/lednet/pytorch/README.md b/cv/semantic_segmentation/lednet/pytorch/README.md
index 06b5817ca..c120ef48f 100644
--- a/cv/semantic_segmentation/lednet/pytorch/README.md
+++ b/cv/semantic_segmentation/lednet/pytorch/README.md
@@ -8,6 +8,12 @@ where two new operations, channel split and shuffle, are utilized in each residu
 cost while maintaining higher segmentation accuracy. On the other hand, an attention pyramid network (APN) is employed
 in the decoder to further lighten the entire network complexity.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/linknet/pytorch/README.md b/cv/semantic_segmentation/linknet/pytorch/README.md
index 73bc80fec..157ce7746 100644
--- a/cv/semantic_segmentation/linknet/pytorch/README.md
+++ b/cv/semantic_segmentation/linknet/pytorch/README.md
@@ -6,6 +6,12 @@ A novel deep neural network architecture which allows it to learn without any si
 parameters. The network uses only 11.5 million parameters and 21.2 GFLOPs for processing an image of resolution
 3x640x360.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/mask2former/pytorch/README.md b/cv/semantic_segmentation/mask2former/pytorch/README.md
index e3493e747..3992bfeb6 100644
--- a/cv/semantic_segmentation/mask2former/pytorch/README.md
+++ b/cv/semantic_segmentation/mask2former/pytorch/README.md
@@ -10,6 +10,12 @@ utilize high-resolution features. It feeds successive feature maps from the pixe
 successive Transformer decoder layers in a round-robin fashion. Finally, we incorporate optimization improvements that
 boost model performance without introducing additional computation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/mobileseg/paddlepaddle/README.md b/cv/semantic_segmentation/mobileseg/paddlepaddle/README.md
index 6074932be..f11679ed2 100644
--- a/cv/semantic_segmentation/mobileseg/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/mobileseg/paddlepaddle/README.md
@@ -5,6 +5,12 @@
 MobileSeg models adopt encoder-decoder architecture and use lightweight models as encoder.
 These semantic segmentation models are designed for mobile and edge devices.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/ocnet/pytorch/README.md b/cv/semantic_segmentation/ocnet/pytorch/README.md
index 564782d98..5175c7dda 100644
--- a/cv/semantic_segmentation/ocnet/pytorch/README.md
+++ b/cv/semantic_segmentation/ocnet/pytorch/README.md
@@ -9,6 +9,12 @@ model dense relations between pixels, focusing on object boundaries and structur
 leveraging object context information makes it particularly effective for complex scene understanding tasks in computer
 vision applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/ocrnet/paddlepaddle/README.md b/cv/semantic_segmentation/ocrnet/paddlepaddle/README.md
index 70f3ed840..0071f2d32 100644
--- a/cv/semantic_segmentation/ocrnet/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/ocrnet/paddlepaddle/README.md
@@ -9,6 +9,12 @@ object regions, OCRNet augments each pixel's representation with contextual info
 approach improves segmentation accuracy, particularly in complex scenes, by better capturing object boundaries and
 contextual relationships between different image elements.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/ocrnet/pytorch/README.md b/cv/semantic_segmentation/ocrnet/pytorch/README.md
index 5aad45ce4..22bfe5541 100644
--- a/cv/semantic_segmentation/ocrnet/pytorch/README.md
+++ b/cv/semantic_segmentation/ocrnet/pytorch/README.md
@@ -9,6 +9,12 @@ object regions, OCRNet augments each pixel's representation with contextual info
 approach improves segmentation accuracy, particularly in complex scenes, by better capturing object boundaries and
 contextual relationships between different image elements.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/pp_humansegv1/paddlepaddle/README.md b/cv/semantic_segmentation/pp_humansegv1/paddlepaddle/README.md
index 415bb4b00..c5cd3fd9a 100644
--- a/cv/semantic_segmentation/pp_humansegv1/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/pp_humansegv1/paddlepaddle/README.md
@@ -9,6 +9,12 @@ fine-tuning for enhanced performance. PP-HumanSegV1 is particularly valuable for
 replacement, portrait snapshot, and barrage penetration, providing high-quality segmentation results with minimal
 computational requirements.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/pp_humansegv2/paddlepaddle/README.md b/cv/semantic_segmentation/pp_humansegv2/paddlepaddle/README.md
index 1d26fd47c..cd2d3400d 100644
--- a/cv/semantic_segmentation/pp_humansegv2/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/pp_humansegv2/paddlepaddle/README.md
@@ -8,6 +8,12 @@ compared to its predecessor. The model supports zero-cost deployment for immedia
 fine-tuning for better performance. PP-HumanSegV2 is particularly effective for applications like video background
 replacement and portrait segmentation, delivering high-quality results with optimized computational efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/pp_liteseg/paddlepaddle/README.md b/cv/semantic_segmentation/pp_liteseg/paddlepaddle/README.md
index 463d8e4ab..59c3cf7fc 100644
--- a/cv/semantic_segmentation/pp_liteseg/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/pp_liteseg/paddlepaddle/README.md
@@ -8,6 +8,12 @@ representations, this model proposes a Unified Attention Fusion Module (UAFM), w
 channel attention to produce a weight and then fuses the input features with the weight. Moreover, a Simple Pyramid
 Pooling Module (SPPM) is proposed to aggregate global context with low computation cost.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/psanet/pytorch/README.md b/cv/semantic_segmentation/psanet/pytorch/README.md
index 18c73228f..6d13e56c0 100644
--- a/cv/semantic_segmentation/psanet/pytorch/README.md
+++ b/cv/semantic_segmentation/psanet/pytorch/README.md
@@ -8,6 +8,12 @@ propagation in bi-direction for scene parsing is enabled. Information at other p
 prediction of the current position and vice versa, information at the current position can be distributed to assist the
 prediction of other ones.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/pspnet/pytorch/README.md b/cv/semantic_segmentation/pspnet/pytorch/README.md
index 7606a31ad..d1f5b6ddb 100644
--- a/cv/semantic_segmentation/pspnet/pytorch/README.md
+++ b/cv/semantic_segmentation/pspnet/pytorch/README.md
@@ -8,6 +8,12 @@ achieves state-ofthe-art performance on various datasets. It came first in Image
 VOC 2012 benchmark and Cityscapes benchmark. A single PSPNet yields the new record of mIoU accuracy 85.4% on PASCAL VOC
 2012 and accuracy 80.2% on Cityscapes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/refinenet/pytorch/README.md b/cv/semantic_segmentation/refinenet/pytorch/README.md
index 164645f25..472c9d92c 100644
--- a/cv/semantic_segmentation/refinenet/pytorch/README.md
+++ b/cv/semantic_segmentation/refinenet/pytorch/README.md
@@ -9,6 +9,12 @@ convolutions. The individual components of RefineNet employ residual connections
 which allows for effective end-to-end training. Further, we introduce chained residual pooling, which captures rich
 background context in an efficient manner.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/segnet/pytorch/README.md b/cv/semantic_segmentation/segnet/pytorch/README.md
index b6bf7f1cd..c1d703d6d 100644
--- a/cv/semantic_segmentation/segnet/pytorch/README.md
+++ b/cv/semantic_segmentation/segnet/pytorch/README.md
@@ -10,6 +10,12 @@ of SegNet lies is in the manner in which the decoder upsamples its lower resolut
 the decoder uses pooling indices computed in the max-pooling step of the corresponding encoder to perform non-linear
 upsampling.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/stdc/paddlepaddle/README.md b/cv/semantic_segmentation/stdc/paddlepaddle/README.md
index 134b1de2b..fc533b5c4 100644
--- a/cv/semantic_segmentation/stdc/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/stdc/paddlepaddle/README.md
@@ -9,6 +9,12 @@ information learning in low-level layers. By fusing both low-level and deep feat
 segmentation results with optimized computational efficiency, making it particularly suitable for real-time
 applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/stdc/pytorch/README.md b/cv/semantic_segmentation/stdc/pytorch/README.md
index 6343c3bba..1edb2fea5 100644
--- a/cv/semantic_segmentation/stdc/pytorch/README.md
+++ b/cv/semantic_segmentation/stdc/pytorch/README.md
@@ -9,6 +9,12 @@ information learning in low-level layers. By fusing both low-level and deep feat
 segmentation results with optimized computational efficiency, making it particularly suitable for real-time
 applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/unet++/pytorch/README.md b/cv/semantic_segmentation/unet++/pytorch/README.md
index 07989de11..53bdf0337 100644
--- a/cv/semantic_segmentation/unet++/pytorch/README.md
+++ b/cv/semantic_segmentation/unet++/pytorch/README.md
@@ -8,6 +8,12 @@ encoder and decoder feature maps, making the optimization task easier. By enhanc
 network levels, UNet++ improves segmentation accuracy, particularly in complex medical imaging tasks. Its architecture
 effectively handles the challenges of precise boundary detection and small object segmentation in medical images.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/unet/paddlepaddle/README.md b/cv/semantic_segmentation/unet/paddlepaddle/README.md
index bb2d22014..4df09a4fd 100644
--- a/cv/semantic_segmentation/unet/paddlepaddle/README.md
+++ b/cv/semantic_segmentation/unet/paddlepaddle/README.md
@@ -6,6 +6,12 @@ A network and training strategy that relies on the strong use of data augmentati
 samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding
 path that enables precise localization.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/unet/pytorch/README.md b/cv/semantic_segmentation/unet/pytorch/README.md
index eb00fc6cb..fde9e8b87 100644
--- a/cv/semantic_segmentation/unet/pytorch/README.md
+++ b/cv/semantic_segmentation/unet/pytorch/README.md
@@ -6,6 +6,12 @@ A network and training strategy that relies on the strong use of data augmentati
 samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding
 path that enables precise localization.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/unet3d/pytorch/README.md b/cv/semantic_segmentation/unet3d/pytorch/README.md
index 88f1f3bd8..5821ae231 100644
--- a/cv/semantic_segmentation/unet3d/pytorch/README.md
+++ b/cv/semantic_segmentation/unet3d/pytorch/README.md
@@ -8,6 +8,12 @@ sparsely annotated volumes to produce dense 3D segmentations. The model supports
 segmentation workflows, incorporating on-the-fly elastic deformations for efficient data augmentation. 3D-UNet is
 particularly valuable in medical imaging for tasks requiring precise 3D anatomical structure delineation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/semantic_segmentation/vnet/tensorflow/README.md b/cv/semantic_segmentation/vnet/tensorflow/README.md
index e719fb849..56e9be097 100644
--- a/cv/semantic_segmentation/vnet/tensorflow/README.md
+++ b/cv/semantic_segmentation/vnet/tensorflow/README.md
@@ -8,6 +8,12 @@ architecture incorporates residual connections and volumetric convolutions to ca
 dimensions. VNet's innovative design enables precise segmentation of complex anatomical structures, making it
 particularly valuable in medical imaging tasks such as organ segmentation and tumor detection in volumetric datasets.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/basicvsr++/pytorch/README.md b/cv/super_resolution/basicvsr++/pytorch/README.md
index 2674fbf04..b2b68ba77 100755
--- a/cv/super_resolution/basicvsr++/pytorch/README.md
+++ b/cv/super_resolution/basicvsr++/pytorch/README.md
@@ -9,6 +9,12 @@ including compressed video enhancement, and achieved top results in NTIRE 2021 c
 structure effectively processes entire video sequences, making it a state-of-the-art solution for high-quality video
 upscaling and restoration.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/basicvsr/pytorch/README.md b/cv/super_resolution/basicvsr/pytorch/README.md
index 5c30134bf..e33589a61 100755
--- a/cv/super_resolution/basicvsr/pytorch/README.md
+++ b/cv/super_resolution/basicvsr/pytorch/README.md
@@ -8,6 +8,12 @@ Figure, red and blue colors represent the backward and forward propagations, res
 contain only generic components. S, W and R refer to the flow estimation module, spatial warping module, and residual
 blocks, respectively.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/esrgan/pytorch/README.md b/cv/super_resolution/esrgan/pytorch/README.md
index 957c5b687..1f25e8647 100755
--- a/cv/super_resolution/esrgan/pytorch/README.md
+++ b/cv/super_resolution/esrgan/pytorch/README.md
@@ -1,51 +1,57 @@
-# ESRGAN
-
-## Model Description
-
-ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) is an advanced deep learning model for single image
-super-resolution. It improves upon SRGAN by introducing Residual-in-Residual Dense Blocks (RRDB) without batch
-normalization, relativistic GAN for the discriminator, and enhanced perceptual loss using pre-activation features. These
-innovations enable ESRGAN to generate more realistic textures with fewer artifacts, producing higher-quality upscaled
-images. It achieved first place in the PIRM2018-SR Challenge, demonstrating superior visual quality and more natural
-textures compared to its predecessor.
-
-## Model Preparation
-
-### Prepare Resources
-
-```shell
-# Download DIV2K: https://data.vision.ee.ethz.ch/cvl/DIV2K/ or you can follow this tools/dataset_converters/div2k/README.md
-$ mkdir -p data/DIV2K
-```
-
-### Install Dependencies
-
-```shell
-# Install libGL
-## CentOS
-yum install -y mesa-libGL
-## Ubuntu
-apt install -y libgl1-mesa-glx
-
-git clone https://github.com/open-mmlab/mmagic.git -b v1.2.0 --depth=1
-cd mmagic/
-pip3 install -e . -v
-
-sed -i 's/diffusers.models.unet_2d_condition/diffusers.models.unets.unet_2d_condition/g' mmagic/models/editors/vico/vico_utils.py
-pip install albumentations
-```
-
-## Model Training
-
-```shell
-# One single GPU
-python3 tools/train.py configs/esrgan/esrgan_psnr-x4c64b23g32_1xb16-1000k_div2k.py
-
-# Mutiple GPUs on one machine
-sed -i 's/python /python3 /g' tools/dist_train.sh
-bash tools/dist_train.sh configs/esrgan/esrgan_psnr-x4c64b23g32_1xb16-1000k_div2k.py 8
-```
-
-## References
-
-- [mmagic](https://github.com/open-mmlab/mmagic)
+# ESRGAN
+
+## Model Description
+
+ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) is an advanced deep learning model for single image
+super-resolution. It improves upon SRGAN by introducing Residual-in-Residual Dense Blocks (RRDB) without batch
+normalization, relativistic GAN for the discriminator, and enhanced perceptual loss using pre-activation features. These
+innovations enable ESRGAN to generate more realistic textures with fewer artifacts, producing higher-quality upscaled
+images. It achieved first place in the PIRM2018-SR Challenge, demonstrating superior visual quality and more natural
+textures compared to its predecessor.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
+## Model Preparation
+
+### Prepare Resources
+
+```shell
+# Download DIV2K: https://data.vision.ee.ethz.ch/cvl/DIV2K/ or you can follow this tools/dataset_converters/div2k/README.md
+$ mkdir -p data/DIV2K
+```
+
+### Install Dependencies
+
+```shell
+# Install libGL
+## CentOS
+yum install -y mesa-libGL
+## Ubuntu
+apt install -y libgl1-mesa-glx
+
+git clone https://github.com/open-mmlab/mmagic.git -b v1.2.0 --depth=1
+cd mmagic/
+pip3 install -e . -v
+
+sed -i 's/diffusers.models.unet_2d_condition/diffusers.models.unets.unet_2d_condition/g' mmagic/models/editors/vico/vico_utils.py
+pip install albumentations
+```
+
+## Model Training
+
+```shell
+# One single GPU
+python3 tools/train.py configs/esrgan/esrgan_psnr-x4c64b23g32_1xb16-1000k_div2k.py
+
+# Mutiple GPUs on one machine
+sed -i 's/python /python3 /g' tools/dist_train.sh
+bash tools/dist_train.sh configs/esrgan/esrgan_psnr-x4c64b23g32_1xb16-1000k_div2k.py 8
+```
+
+## References
+
+- [mmagic](https://github.com/open-mmlab/mmagic)
diff --git a/cv/super_resolution/liif/pytorch/README.md b/cv/super_resolution/liif/pytorch/README.md
index f0106f01a..fddaf48fb 100755
--- a/cv/super_resolution/liif/pytorch/README.md
+++ b/cv/super_resolution/liif/pytorch/README.md
@@ -8,6 +8,12 @@ arbitrary resolution representation. LIIF combines 2D deep features with coordin
 images, even at resolutions 30x higher than training data. This approach bridges discrete and continuous image
 representations, outperforming traditional resizing methods and supporting tasks with varying image sizes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/real_basicvsr/pytorch/README.md b/cv/super_resolution/real_basicvsr/pytorch/README.md
index d45dab40b..590aabb4a 100755
--- a/cv/super_resolution/real_basicvsr/pytorch/README.md
+++ b/cv/super_resolution/real_basicvsr/pytorch/README.md
@@ -8,6 +8,12 @@ and artifact suppression. The model introduces a stochastic degradation scheme t
 performance, and emphasizes the use of longer sequences over larger batches for more effective temporal information
 utilization. RealBasicVSR demonstrates superior quality and efficiency in video enhancement tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/ttsr/pytorch/README.md b/cv/super_resolution/ttsr/pytorch/README.md
index 9bc1e608e..809d05163 100755
--- a/cv/super_resolution/ttsr/pytorch/README.md
+++ b/cv/super_resolution/ttsr/pytorch/README.md
@@ -9,6 +9,12 @@ hard-attention for texture transfer, and soft-attention for texture synthesis. T
 transfer through attention mechanisms, allowing for high-quality image reconstruction at various magnification levels
 (1x to 4x).
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/cv/super_resolution/ttvsr/pytorch/README.md b/cv/super_resolution/ttvsr/pytorch/README.md
index bafacc6c3..10e7403f1 100755
--- a/cv/super_resolution/ttvsr/pytorch/README.md
+++ b/cv/super_resolution/ttvsr/pytorch/README.md
@@ -9,6 +9,12 @@ approach improves video super-resolution by better utilizing temporal informatio
 quality upscaled videos. TTVSR demonstrates superior performance in handling complex video sequences while maintaining
 efficient processing capabilities.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/docs/MODEL_TEMPLATE.md b/docs/MODEL_TEMPLATE.md
index 45c826646..dde05d2a5 100644
--- a/docs/MODEL_TEMPLATE.md
+++ b/docs/MODEL_TEMPLATE.md
@@ -8,10 +8,10 @@ A brief introduction about this model.
 
 ## Supported Environments
 
-| Iluvatar GPU | IXUCA Version |
-|--------------|---------------|
-| BI-V100      | 3.0.0         |
-| BI-V150      | 4.2.0         |
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.2.0     |  25.03  |
+| BI-V100 | 3.2.0     |  23.03  |
 
 ## Model Preparation
 
diff --git a/hpc/molecular_dynamics/water_se_e2_a/tensorflow/README.md b/hpc/molecular_dynamics/water_se_e2_a/tensorflow/README.md
index 7109b71e3..fdeb13b34 100644
--- a/hpc/molecular_dynamics/water_se_e2_a/tensorflow/README.md
+++ b/hpc/molecular_dynamics/water_se_e2_a/tensorflow/README.md
@@ -4,7 +4,17 @@ The notation of se_e2_a is short for the Deep Potential Smooth Edition (DeepPot-
 Note that it is sometimes called a “two-atom embedding descriptor” which means the input of the embedding net is atomic distances. The descriptor does encode multi-body information (both angular and radial information of neighboring atoms).
 In this example, we will train a DeepPot-SE model for a water system. A complete training input script of this example can be found in the directory.
 
-## Step 1: Installation
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.09  |
+
+## Model Preparation
+
+### Prepare Resources
+
+### Install Dependencies
 ```
 apt install git
 git clone --recursive -b v2.2.2 https://github.com/deepmodeling/deepmd-kit.git deepmd-kit
@@ -12,7 +22,7 @@ pip3 install numpy==1.22.3
 pip3 install deepmd-kit[gpu,cu10,lmp,ipi]==2.2.2
 ```
 
-### Install the DeePMD-kit’s Python interface
+#### Install the DeePMD-kit’s Python interface
 Visit Iluvatar Corex official website - Resource Center page (https://support.iluvatar.com/#/DocumentCentre?id=1&nameCenter=2&productId=381380977957597184) to obtain the Linux version software stack offline installation package. If you already have an account, click the "Login" button at the upper right corner. If you do not have an account, click the "Login" button and select "Register" to apply for an account. And then download BI_SDK3.1.0 (4.57GB).
 ```
 pip3 install tensorflow-2.6.5+corex.3.1.0-cp38-cp38-linux_x86_64.whl
@@ -20,7 +30,7 @@ sed -i '473s/np.float64/np.float32/' deepmd/env.py
 pip3 install .
 ```
 
-### Install the DeePMD-kit’s C++ interface
+#### Install the DeePMD-kit’s C++ interface
 ```
 deepmd_source_dir=`pwd`
 cd $deepmd_source_dir/source
@@ -31,14 +41,14 @@ make -j4
 make install
 ```
 
-### Install from pre-compiled C library
+#### Install from pre-compiled C library
 ```
 cmake -DDEEPMD_C_ROOT=./libdeepmd_c -DCMAKE_INSTALL_PREFIX=$deepmd_root ..
 make -j8
 make install
 ```
 
-### Install LAMMPS
+#### Install LAMMPS
 ```
 cd $deepmd_source_dir
 wget https://github.com/lammps/lammps/archive/stable_23Jun2022_update4.tar.gz
@@ -53,14 +63,14 @@ make -j4
 make install
 ```
 
-### Install i-PI
+#### Install i-PI
 ```
 cd ../..
 pip3 install -U i-PI
 pip3 install pytest
 ```
 
-## Step 2: Training
+## Model Training
 
 ### One single GPU
 ```
@@ -70,12 +80,12 @@ export TF_ENABLE_DEPRECATION_WARNINGS=1
 DP_INTERFACE_PREC=low dp train input.json
 ```
 
-## Results
+## Model Results
 | GPU         | average training     |
 | ----------- | -------------------- |
 | 1 card      | 0.0325 s/batch       |
 
 
-## Reference
+## References
 https://github.com/deepmodeling/deepmd-kit#about-deepmd-kit
 
diff --git a/multimodal/contrastive_learning/clip/pytorch/README.md b/multimodal/contrastive_learning/clip/pytorch/README.md
index 330516e0d..5c26ef895 100644
--- a/multimodal/contrastive_learning/clip/pytorch/README.md
+++ b/multimodal/contrastive_learning/clip/pytorch/README.md
@@ -8,6 +8,12 @@ encoder and a text encoder to predict the correct pairings of a batch of (image,
 the learned text encoder synthesizes a zero-shot linear classifier by embedding the names or descriptions of the target
 dataset’s classes.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/controlnet/pytorch/README.md b/multimodal/diffusion_model/controlnet/pytorch/README.md
index d42250228..a040b3d8e 100644
--- a/multimodal/diffusion_model/controlnet/pytorch/README.md
+++ b/multimodal/diffusion_model/controlnet/pytorch/README.md
@@ -12,6 +12,12 @@ Stable diffusion is trained on billions of images, and it already knows what is
 
 But it does not know the meaning of that "Control Image (Source Image)". Our target is to let it know.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/ddpm/pytorch/README.md b/multimodal/diffusion_model/ddpm/pytorch/README.md
index 6fdf2649c..e7a3bd272 100644
--- a/multimodal/diffusion_model/ddpm/pytorch/README.md
+++ b/multimodal/diffusion_model/ddpm/pytorch/README.md
@@ -8,6 +8,12 @@ adding Gaussian noise to data during training and then learning to reverse this
 to generate high-quality samples by starting from random noise and iteratively refining it. DDPMs have shown impressive
 results in image generation, offering stable training and producing diverse, realistic outputs.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/stable-diffusion-1.4/pytorch/README.md b/multimodal/diffusion_model/stable-diffusion-1.4/pytorch/README.md
index 10af10f45..4c4d39faa 100644
--- a/multimodal/diffusion_model/stable-diffusion-1.4/pytorch/README.md
+++ b/multimodal/diffusion_model/stable-diffusion-1.4/pytorch/README.md
@@ -8,6 +8,12 @@ images by operating in a compressed latent space. The model leverages a frozen C
 and process input prompts. With its 860M UNet and 123M text encoder, Stable Diffusion achieves remarkable results while
 maintaining computational efficiency, making it accessible for users with GPUs having at least 4GB VRAM.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/multimodal/diffusion_model/stable-diffusion-1.5/pytorch/README.md b/multimodal/diffusion_model/stable-diffusion-1.5/pytorch/README.md
index 19099d741..9704f14f1 100644
--- a/multimodal/diffusion_model/stable-diffusion-1.5/pytorch/README.md
+++ b/multimodal/diffusion_model/stable-diffusion-1.5/pytorch/README.md
@@ -9,6 +9,12 @@ while maintaining exceptional visual quality. With its ability to interpret dive
 corresponding images, Stable Diffusion 1.5 has become a powerful tool for creative applications, AI-assisted design, and
 visual content generation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/stable-diffusion-2.1/pytorch/README.md b/multimodal/diffusion_model/stable-diffusion-2.1/pytorch/README.md
index a2b3c9093..63ddc390b 100644
--- a/multimodal/diffusion_model/stable-diffusion-2.1/pytorch/README.md
+++ b/multimodal/diffusion_model/stable-diffusion-2.1/pytorch/README.md
@@ -9,6 +9,12 @@ understanding compared to earlier versions. The model operates efficiently in a
 accessible for various applications while maintaining exceptional visual fidelity. Stable Diffusion 2.1 has become a
 powerful tool for creative professionals and AI enthusiasts alike.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/stable-diffusion-3/pytorch/README.md b/multimodal/diffusion_model/stable-diffusion-3/pytorch/README.md
index 6148fcc52..7606bac7f 100644
--- a/multimodal/diffusion_model/stable-diffusion-3/pytorch/README.md
+++ b/multimodal/diffusion_model/stable-diffusion-3/pytorch/README.md
@@ -9,6 +9,12 @@ prompt comprehension. With its ability to generate highly detailed and contextua
 pushes the boundaries of AI-assisted creativity. The model maintains efficient processing through its latent space
 operations while delivering state-of-the-art results in image synthesis and generation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/diffusion_model/stable-diffusion-xl/pytorch/README.md b/multimodal/diffusion_model/stable-diffusion-xl/pytorch/README.md
index 76a3fe59d..cab188b62 100644
--- a/multimodal/diffusion_model/stable-diffusion-xl/pytorch/README.md
+++ b/multimodal/diffusion_model/stable-diffusion-xl/pytorch/README.md
@@ -9,6 +9,12 @@ techniques, Stable Diffusion XL excels at producing photorealistic and artistic
 diversity. The model's ability to interpret complex prompts and generate corresponding images makes it a valuable tool
 for creative professionals, designers, and AI enthusiasts.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/vision-language_model/blip/pytorch/README.md b/multimodal/vision-language_model/blip/pytorch/README.md
index b00ddc071..bc829e712 100755
--- a/multimodal/vision-language_model/blip/pytorch/README.md
+++ b/multimodal/vision-language_model/blip/pytorch/README.md
@@ -8,6 +8,12 @@ visual comprehension and text generation. It employs a unique bootstrapping mech
 web-sourced image-text pairs, improving the quality of training data. This approach enables BLIP to achieve superior
 performance in tasks like image captioning, visual question answering, and multimodal understanding.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.1     |  24.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/vision-language_model/l-verse/pytorch/README.md b/multimodal/vision-language_model/l-verse/pytorch/README.md
index 0b2191f9e..af09c96c8 100644
--- a/multimodal/vision-language_model/l-verse/pytorch/README.md
+++ b/multimodal/vision-language_model/l-verse/pytorch/README.md
@@ -9,6 +9,12 @@ without requiring fine-tuning or additional frameworks. Its AugVAE component ach
 reconstruction, while BiART effectively distinguishes between conditional references and generation targets. L-Verse
 demonstrates impressive results in multimodal tasks, particularly on MS-COCO Captions dataset.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/vision-language_model/llava-1.5/pytorch/README.md b/multimodal/vision-language_model/llava-1.5/pytorch/README.md
index 3cb47c183..df97f7e87 100644
--- a/multimodal/vision-language_model/llava-1.5/pytorch/README.md
+++ b/multimodal/vision-language_model/llava-1.5/pytorch/README.md
@@ -6,6 +6,12 @@ LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-gener
 instruction-following data. It is an auto-regressive language model, based on the transformer
 architecture.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.12  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/multimodal/vision-language_model/moe-llava-phi2-2.7b/pytorch/README.md b/multimodal/vision-language_model/moe-llava-phi2-2.7b/pytorch/README.md
index 0a98afec1..7272152bb 100644
--- a/multimodal/vision-language_model/moe-llava-phi2-2.7b/pytorch/README.md
+++ b/multimodal/vision-language_model/moe-llava-phi2-2.7b/pytorch/README.md
@@ -8,6 +8,12 @@ information. The model leverages expert networks to specialize in different aspe
 enabling more accurate and context-aware responses. MoE-LLaVA is particularly effective in applications requiring
 complex reasoning across visual and linguistic domains, such as image captioning and visual question answering.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.2.0     |  25.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/vision-language_model/moe-llava-qwen-1.8b/pytorch/README.md b/multimodal/vision-language_model/moe-llava-qwen-1.8b/pytorch/README.md
index bf216ff03..36ef33bc7 100644
--- a/multimodal/vision-language_model/moe-llava-qwen-1.8b/pytorch/README.md
+++ b/multimodal/vision-language_model/moe-llava-qwen-1.8b/pytorch/README.md
@@ -9,6 +9,12 @@ information. The model leverages expert networks to specialize in different aspe
 enabling more accurate and context-aware responses. MoE-LLaVA is particularly effective in applications requiring
 complex reasoning across visual and linguistic domains, such as image captioning and visual question answering.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.2.0     |  25.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/multimodal/vision-language_model/moe-llava-stablelm-1.6b/pytorch/README.md b/multimodal/vision-language_model/moe-llava-stablelm-1.6b/pytorch/README.md
index 4c34f8dcf..a40dd6207 100644
--- a/multimodal/vision-language_model/moe-llava-stablelm-1.6b/pytorch/README.md
+++ b/multimodal/vision-language_model/moe-llava-stablelm-1.6b/pytorch/README.md
@@ -8,6 +8,12 @@ information. The model leverages expert networks to specialize in different aspe
 enabling more accurate and context-aware responses. MoE-LLaVA is particularly effective in applications requiring
 complex reasoning across visual and linguistic domains, such as image captioning and visual question answering.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.2.0     |  25.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/cloze_test/glm/pytorch/GLMForMultiTokenCloze/README.md b/nlp/cloze_test/glm/pytorch/README.md
similarity index 85%
rename from nlp/cloze_test/glm/pytorch/GLMForMultiTokenCloze/README.md
rename to nlp/cloze_test/glm/pytorch/README.md
index 40a178be7..63bfb2b48 100644
--- a/nlp/cloze_test/glm/pytorch/GLMForMultiTokenCloze/README.md
+++ b/nlp/cloze_test/glm/pytorch/README.md
@@ -9,6 +9,12 @@ including NLU, conditional generation, and unconditional generation. With its ab
 tasks through adjustable blank configurations, GLM outperforms specialized models like BERT, T5, and GPT while
 maintaining efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/dialogue_generation/cpm/pytorch/README.md b/nlp/dialogue_generation/cpm/pytorch/README.md
index 29ab6f0b9..d2951b139 100644
--- a/nlp/dialogue_generation/cpm/pytorch/README.md
+++ b/nlp/dialogue_generation/cpm/pytorch/README.md
@@ -9,6 +9,12 @@ enables effective few-shot and zero-shot learning capabilities, making it partic
 processing. As one of the largest Chinese language models, CPM significantly advances the state of Chinese NLP research
 and applications.s
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/bart_fairseq/pytorch/README.md b/nlp/language_model/bart_fairseq/pytorch/README.md
index 41c1943c5..3c0a77698 100644
--- a/nlp/language_model/bart_fairseq/pytorch/README.md
+++ b/nlp/language_model/bart_fairseq/pytorch/README.md
@@ -9,6 +9,12 @@ allows it to effectively handle both understanding and generation tasks, making
 applications. BART has demonstrated state-of-the-art performance on benchmarks like XSum, CNN/Daily Mail, and GLUE,
 showcasing its robust capabilities in text transformation and comprehension.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/bert/mindspore/README.md b/nlp/language_model/bert/mindspore/README.md
index 3b6bbd940..e13c25aef 100644
--- a/nlp/language_model/bert/mindspore/README.md
+++ b/nlp/language_model/bert/mindspore/README.md
@@ -8,6 +8,12 @@ context from both directions in text. Pretrained using Masked Language Modeling
 tasks, BERT achieves state-of-the-art results across various NLP tasks through fine-tuning. Its ability to understand
 deep contextual relationships in text has made it a fundamental model in modern NLP research and applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/bert/paddlepaddle/README.md b/nlp/language_model/bert/paddlepaddle/README.md
index 271858a91..f852aed62 100644
--- a/nlp/language_model/bert/paddlepaddle/README.md
+++ b/nlp/language_model/bert/paddlepaddle/README.md
@@ -8,6 +8,12 @@ context from both directions in text. Pretrained using Masked Language Modeling
 tasks, BERT achieves state-of-the-art results across various NLP tasks through fine-tuning. Its ability to understand
 deep contextual relationships in text has made it a fundamental model in modern NLP research and applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/bert/pytorch/README.md b/nlp/language_model/bert/pytorch/README.md
index bec27fdfd..771aa96f9 100644
--- a/nlp/language_model/bert/pytorch/README.md
+++ b/nlp/language_model/bert/pytorch/README.md
@@ -8,6 +8,12 @@ context from both directions in text. Pretrained using Masked Language Modeling
 tasks, BERT achieves state-of-the-art results across various NLP tasks through fine-tuning. Its ability to understand
 deep contextual relationships in text has made it a fundamental model in modern NLP research and applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/bert/tensorflow/base/README.md b/nlp/language_model/bert/tensorflow/README.md
similarity index 92%
rename from nlp/language_model/bert/tensorflow/base/README.md
rename to nlp/language_model/bert/tensorflow/README.md
index c487108de..e2c7074f7 100644
--- a/nlp/language_model/bert/tensorflow/base/README.md
+++ b/nlp/language_model/bert/tensorflow/README.md
@@ -1,68 +1,74 @@
-# BERT Pretraining
-
-## Model Description
-
-BERT (Bidirectional Encoder Representations from Transformers) is a groundbreaking language model that revolutionized
-natural language processing. It employs a transformer architecture with bidirectional attention, enabling it to capture
-context from both directions in text. Pretrained using Masked Language Modeling (MLM) and Next Sentence Prediction (NSP)
-tasks, BERT achieves state-of-the-art results across various NLP tasks through fine-tuning. Its ability to understand
-deep contextual relationships in text has made it a fundamental model in modern NLP research and applications.
-
-## Model Preparation
-
-### Prepare Resources
-
-This [Google Drive location](https://drive.google.com/drive/folders/1oQF4diVHNPCclykwdvQJw8n_VIWwV0PT) contains the
-following.  
-You need to download tf1_ckpt folde , vocab.txt and bert_config.json into one file named bert_pretrain_ckpt_tf
-
-```sh
-bert_pretrain_ckpt_tf: contains checkpoint files
-    model.ckpt-28252.data-00000-of-00001
-    model.ckpt-28252.index
-    model.ckpt-28252.meta
-    vocab.txt
-    bert_config.json
-```
-
-[Download and preprocess datasets](https://github.com/mlcommons/training/tree/master/language_model/tensorflow/bert#generate-the-tfrecords-for-wiki-dataset)
-You need to make a file named  bert_pretrain_tf_records and store the results above.
-tips: you can git clone this repo in other place ,we need the bert_pretrain_tf_records results here.
-
-### Install Dependencies
-
-```shell
-bash init_tf.sh
-wget https://download.open-mpi.org/release/open-mpi/v4.0/openmpi-4.0.7.tar.gz
-tar xf openmpi-4.0.7.tar.gz
-cd openmpi-4.0.7/
-./configure --prefix=/usr/local/bin --with-orte
-make -j4 && make install
-export LD_LIBRARY_PATH=/usr/local/lib/:$LD_LIBRARY_PATH
-```
-
-## Model Training
-
-```shell
-# Training on single card
-bash run_1card_FPS.sh --input_files_dir=/path/to/bert_pretrain_tf_records/train_data \
-        --init_checkpoint=/path/to/bert_pretrain_ckpt_tf/model.ckpt-28252 \
-        --eval_files_dir=/path/to/bert_pretrain_tf_records/eval_data \
-        --train_batch_size=6 \
-        --bert_config_file=/path/to/bert_pretrain_ckpt_tf/bert_config.json
-
-# Training on mutil-cards
-export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
-export IX_NUM_CUDA_VISIBLE_DEVICES=8
-bash run_multi_card_FPS.sh --input_files_dir=/path/to/bert_pretrain_tf_records/train_data \
-        --init_checkpoint=/path/to/bert_pretrain_ckpt_tf/model.ckpt-28252 \
-        --eval_files_dir=/path/to/bert_pretrain_tf_records/eval_data \
-        --train_batch_size=6 \
-        --bert_config_file=/path/to/bert_pretrain_ckpt_tf/bert_config.json
-```
-
-## Model Results
-
-| Model            | GPUs       | acc      | fps      |
-|------------------|------------|----------|----------|
-| BERT Pretraining | BI-V100 x8 | 0.424126 | 0.267241 |
+# BERT Pretraining
+
+## Model Description
+
+BERT (Bidirectional Encoder Representations from Transformers) is a groundbreaking language model that revolutionized
+natural language processing. It employs a transformer architecture with bidirectional attention, enabling it to capture
+context from both directions in text. Pretrained using Masked Language Modeling (MLM) and Next Sentence Prediction (NSP)
+tasks, BERT achieves state-of-the-art results across various NLP tasks through fine-tuning. Its ability to understand
+deep contextual relationships in text has made it a fundamental model in modern NLP research and applications.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
+## Model Preparation
+
+### Prepare Resources
+
+This [Google Drive location](https://drive.google.com/drive/folders/1oQF4diVHNPCclykwdvQJw8n_VIWwV0PT) contains the
+following.  
+You need to download tf1_ckpt folde , vocab.txt and bert_config.json into one file named bert_pretrain_ckpt_tf
+
+```sh
+bert_pretrain_ckpt_tf: contains checkpoint files
+    model.ckpt-28252.data-00000-of-00001
+    model.ckpt-28252.index
+    model.ckpt-28252.meta
+    vocab.txt
+    bert_config.json
+```
+
+[Download and preprocess datasets](https://github.com/mlcommons/training/tree/master/language_model/tensorflow/bert#generate-the-tfrecords-for-wiki-dataset)
+You need to make a file named  bert_pretrain_tf_records and store the results above.
+tips: you can git clone this repo in other place ,we need the bert_pretrain_tf_records results here.
+
+### Install Dependencies
+
+```shell
+bash init_tf.sh
+wget https://download.open-mpi.org/release/open-mpi/v4.0/openmpi-4.0.7.tar.gz
+tar xf openmpi-4.0.7.tar.gz
+cd openmpi-4.0.7/
+./configure --prefix=/usr/local/bin --with-orte
+make -j4 && make install
+export LD_LIBRARY_PATH=/usr/local/lib/:$LD_LIBRARY_PATH
+```
+
+## Model Training
+
+```shell
+# Training on single card
+bash run_1card_FPS.sh --input_files_dir=/path/to/bert_pretrain_tf_records/train_data \
+        --init_checkpoint=/path/to/bert_pretrain_ckpt_tf/model.ckpt-28252 \
+        --eval_files_dir=/path/to/bert_pretrain_tf_records/eval_data \
+        --train_batch_size=6 \
+        --bert_config_file=/path/to/bert_pretrain_ckpt_tf/bert_config.json
+
+# Training on mutil-cards
+export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
+export IX_NUM_CUDA_VISIBLE_DEVICES=8
+bash run_multi_card_FPS.sh --input_files_dir=/path/to/bert_pretrain_tf_records/train_data \
+        --init_checkpoint=/path/to/bert_pretrain_ckpt_tf/model.ckpt-28252 \
+        --eval_files_dir=/path/to/bert_pretrain_tf_records/eval_data \
+        --train_batch_size=6 \
+        --bert_config_file=/path/to/bert_pretrain_ckpt_tf/bert_config.json
+```
+
+## Model Results
+
+| Model            | GPUs       | acc      | fps      |
+|------------------|------------|----------|----------|
+| BERT Pretraining | BI-V100 x8 | 0.424126 | 0.267241 |
diff --git a/nlp/language_model/roberta_fairseq/pytorch/README.md b/nlp/language_model/roberta_fairseq/pytorch/README.md
index c1cf78ecc..96c23ba70 100644
--- a/nlp/language_model/roberta_fairseq/pytorch/README.md
+++ b/nlp/language_model/roberta_fairseq/pytorch/README.md
@@ -9,6 +9,12 @@ tasks. By training on longer sequences and optimizing the training procedure, Ro
 understanding capabilities compared to its predecessor, making it a powerful tool for natural language processing
 applications.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/language_model/xlnet/paddlepaddle/README.md b/nlp/language_model/xlnet/paddlepaddle/README.md
index fc3f77559..ab195253a 100644
--- a/nlp/language_model/xlnet/paddlepaddle/README.md
+++ b/nlp/language_model/xlnet/paddlepaddle/README.md
@@ -9,6 +9,12 @@ Additionally, it incorporates Transformer-XL architecture, which handles long-ra
 recurrence and relative positional encoding. XLNet achieves state-of-the-art performance across various NLP tasks by
 leveraging these innovative techniques.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/ner/bert/pytorch/README.md b/nlp/ner/bert/pytorch/README.md
index 2a26ac4ac..558111d5c 100644
--- a/nlp/ner/bert/pytorch/README.md
+++ b/nlp/ner/bert/pytorch/README.md
@@ -9,6 +9,12 @@ recognition accuracy compared to traditional methods. BERT NER's ability to unde
 makes it particularly effective for complex text analysis tasks in various domains, including information extraction and
 text mining.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/question_answering/bert/pytorch/README.md b/nlp/question_answering/bert/pytorch/README.md
index 69be503c1..23b31fa41 100644
--- a/nlp/question_answering/bert/pytorch/README.md
+++ b/nlp/question_answering/bert/pytorch/README.md
@@ -8,6 +8,12 @@ attention mechanism. The model is trained to predict the start and end positions
 demonstrating exceptional performance in comprehension tasks. BERT SQuAD's ability to understand context and
 relationships between words makes it particularly effective for complex question answering scenarios in various domains.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/text_classification/bert/pytorch/README.md b/nlp/text_classification/bert/pytorch/README.md
index 3d114e84b..1caa9b248 100644
--- a/nlp/text_classification/bert/pytorch/README.md
+++ b/nlp/text_classification/bert/pytorch/README.md
@@ -8,6 +8,12 @@ understanding context. By leveraging BERT's bidirectional attention mechanism, i
 linguistic nuances and relationships between text segments. This makes BERT WNLI particularly valuable for tasks
 requiring deep comprehension of sentence structure and meaning, such as coreference resolution and textual entailment.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/nlp/text_correction/ernie/paddlepaddle/README.md b/nlp/text_correction/ernie/paddlepaddle/README.md
index c88e03717..df22f35b1 100644
--- a/nlp/text_correction/ernie/paddlepaddle/README.md
+++ b/nlp/text_correction/ernie/paddlepaddle/README.md
@@ -8,6 +8,12 @@ sources, such as structured knowledge graphs, and by integrating multiple lingui
 semantics, and common sense. The model achieves this by using a knowledge-enhanced pre-training approach, which helps
 ERNIE better understand and generate more accurate and contextually aware language representations.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/nlp/text_summarisation/bert/pytorch/README.md b/nlp/text_summarisation/bert/pytorch/README.md
index 273952a7b..3965ea12b 100644
--- a/nlp/text_summarisation/bert/pytorch/README.md
+++ b/nlp/text_summarisation/bert/pytorch/README.md
@@ -9,6 +9,12 @@ informative summaries while preserving the original meaning. BERT summarization
 applications requiring efficient information extraction and condensation, such as news aggregation, document analysis,
 and content curation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.03  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/nlp/translation/convolutional_fairseq/pytorch/README.md b/nlp/translation/convolutional_fairseq/pytorch/README.md
index f56784486..c3d1f5555 100644
--- a/nlp/translation/convolutional_fairseq/pytorch/README.md
+++ b/nlp/translation/convolutional_fairseq/pytorch/README.md
@@ -1,69 +1,75 @@
-# Convolutional
-
-## Model Description
-
-Convolutional translation models leverage convolutional neural networks (CNNs) for machine translation tasks, offering
-an alternative to traditional RNN-based approaches. These models process input sequences through multiple convolutional
-layers, capturing local patterns and hierarchical features in the text. By using stacked convolutions with gated linear
-units, they effectively model long-range dependencies while maintaining computational efficiency. Convolutional
-translation models are particularly advantageous for parallel processing and handling large-scale translation tasks,
-demonstrating competitive performance in sequence-to-sequence learning scenarios with reduced training time compared to
-recurrent architectures.
-
-## Model Preparation
-
-### Prepare Resources
-
-```bash
-# Go to "toolbox/Fairseq" directory in root path
-cd ../../../../toolbox/Fairseq/
-
-cd fairseq/examples/translation/
-bash prepare-wmt14en2de.sh
-cd ../..
-
-TEXT=examples/translation/wmt17_en_de
-fairseq-preprocess \
-    --source-lang en --target-lang de \
-    --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \
-    --destdir data-bin/wmt17_en_de --thresholdtgt 0 --thresholdsrc 0 \
-    --workers 20
-```
-
-### Install Dependencies
-
-Convolutional model is using Fairseq toolbox. Before you run this model, you need to setup Fairseq first.
-
-```bash
-bash install_toolbox_fairseq.sh
-```
-
-## Model Training
-
-```bash
-# Train
-mkdir -p checkpoints/fconv_wmt_en_de
-fairseq-train data-bin/wmt17_en_de --arch fconv_wmt_en_de \
-                                   --max-epoch 100 \
-                                   --dropout 0.2 \
-                                   --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
-                                   --optimizer nag --clip-norm 0.1 \
-                                   --lr 0.5 --lr-scheduler fixed --force-anneal 50 \
-                                   --max-tokens 4000 \
-                                   --no-epoch-checkpoints \
-                                   --save-dir checkpoints/fconv_wmt_en_de
-
-# Evaluate
-fairseq-generate data-bin/wmt17_en_de --path checkpoints/fconv_wmt_en_de/checkpoint_best.pt \
-                                      --beam 5 --remove-bpe
-```
-
-## Model Results
-
-| Model         | GPUs       | QPS     | Train Epochs | Evaluate_Bleu |
-|---------------|------------|---------|--------------|---------------|
-| Convolutional | BI-V100 x8 | 1650.49 | 100          | 25.55         |
-
-## References
-
-- [Fairseq](https://github.com/facebookresearch/fairseq/tree/v0.10.2)
+# Convolutional
+
+## Model Description
+
+Convolutional translation models leverage convolutional neural networks (CNNs) for machine translation tasks, offering
+an alternative to traditional RNN-based approaches. These models process input sequences through multiple convolutional
+layers, capturing local patterns and hierarchical features in the text. By using stacked convolutions with gated linear
+units, they effectively model long-range dependencies while maintaining computational efficiency. Convolutional
+translation models are particularly advantageous for parallel processing and handling large-scale translation tasks,
+demonstrating competitive performance in sequence-to-sequence learning scenarios with reduced training time compared to
+recurrent architectures.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
+## Model Preparation
+
+### Prepare Resources
+
+```bash
+# Go to "toolbox/Fairseq" directory in root path
+cd ../../../../toolbox/Fairseq/
+
+cd fairseq/examples/translation/
+bash prepare-wmt14en2de.sh
+cd ../..
+
+TEXT=examples/translation/wmt17_en_de
+fairseq-preprocess \
+    --source-lang en --target-lang de \
+    --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \
+    --destdir data-bin/wmt17_en_de --thresholdtgt 0 --thresholdsrc 0 \
+    --workers 20
+```
+
+### Install Dependencies
+
+Convolutional model is using Fairseq toolbox. Before you run this model, you need to setup Fairseq first.
+
+```bash
+bash install_toolbox_fairseq.sh
+```
+
+## Model Training
+
+```bash
+# Train
+mkdir -p checkpoints/fconv_wmt_en_de
+fairseq-train data-bin/wmt17_en_de --arch fconv_wmt_en_de \
+                                   --max-epoch 100 \
+                                   --dropout 0.2 \
+                                   --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
+                                   --optimizer nag --clip-norm 0.1 \
+                                   --lr 0.5 --lr-scheduler fixed --force-anneal 50 \
+                                   --max-tokens 4000 \
+                                   --no-epoch-checkpoints \
+                                   --save-dir checkpoints/fconv_wmt_en_de
+
+# Evaluate
+fairseq-generate data-bin/wmt17_en_de --path checkpoints/fconv_wmt_en_de/checkpoint_best.pt \
+                                      --beam 5 --remove-bpe
+```
+
+## Model Results
+
+| Model         | GPUs       | QPS     | Train Epochs | Evaluate_Bleu |
+|---------------|------------|---------|--------------|---------------|
+| Convolutional | BI-V100 x8 | 1650.49 | 100          | 25.55         |
+
+## References
+
+- [Fairseq](https://github.com/facebookresearch/fairseq/tree/v0.10.2)
diff --git a/nlp/translation/t5/pytorch/README.md b/nlp/translation/t5/pytorch/README.md
index b752f2303..336eb1684 100644
--- a/nlp/translation/t5/pytorch/README.md
+++ b/nlp/translation/t5/pytorch/README.md
@@ -8,6 +8,12 @@ text generation problem. This allows T5 to use the same architecture and trainin
 By converting inputs and outputs into text sequences, T5 demonstrates strong performance across multiple benchmarks
 while maintaining a consistent and scalable approach to natural language processing tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/nlp/translation/transformer/paddlepaddle/README.md b/nlp/translation/transformer/paddlepaddle/README.md
index 93ae1df25..e2df96e62 100644
--- a/nlp/translation/transformer/paddlepaddle/README.md
+++ b/nlp/translation/transformer/paddlepaddle/README.md
@@ -10,6 +10,12 @@ multi-head attention and position-wise feed-forward networks. Transformers have
 state-of-the-art models like BERT, GPT, and T5, driving advancements in machine translation, text generation, and other
 NLP tasks.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/nlp/translation/transformer_fairseq/pytorch/README.md b/nlp/translation/transformer_fairseq/pytorch/README.md
index b7a3f7da0..94ecbd4f4 100644
--- a/nlp/translation/transformer_fairseq/pytorch/README.md
+++ b/nlp/translation/transformer_fairseq/pytorch/README.md
@@ -1,75 +1,81 @@
-# Transformer
-
-## Model Description
-
-The Transformer model revolutionized natural language processing with its attention-based architecture, eliminating the
-need for recurrent connections. It employs self-attention mechanisms to process input sequences in parallel, capturing
-long-range dependencies more effectively than previous models. Transformers excel in tasks like translation, text
-generation, and summarization by dynamically weighting the importance of different words in a sequence. Their parallel
-processing capability enables faster training and better scalability, making them the foundation for state-of-the-art
-language models like BERT and GPT.
-
-## Model Preparation
-
-### Prepare Resources
-
-```bash
-# Go to "toolbox/Fairseq" directory in root path
-cd ../../../../toolbox/Fairseq/
-
-cd fairseq/examples/translation/
-bash prepare-iwslt14.sh
-cd ../..
-
-TEXT=examples/translation/iwslt14.tokenized.de-en
-fairseq-preprocess --source-lang de --target-lang en \
-    --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \
-    --destdir data-bin/iwslt14.tokenized.de-en \
-    --workers 20
-```
-
-### Install Dependencies
-
-Transformer model is using Fairseq toolbox. Before you run this model, you need to setup Fairseq first.
-
-```bash
-bash install_toolbox_fairseq.sh
-```
-
-## Model Training
-
-```bash
-# Train
-mkdir -p checkpoints/transformer
-fairseq-train data-bin/iwslt14.tokenized.de-en \
-    --arch transformer_iwslt_de_en --share-decoder-input-output-embed \
-    --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \
-    --lr 5e-4 --lr-scheduler inverse_sqrt --warmup-updates 4000 \
-    --dropout 0.3 --weight-decay 0.0001 \
-    --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
-    --max-tokens 4096 \
-    --max-epoch 100 \
-    --eval-bleu \
-    --eval-bleu-args '{"beam": 5, "max_len_a": 1.2, "max_len_b": 10}' \
-    --eval-bleu-detok moses \
-    --eval-bleu-remove-bpe \
-    --eval-bleu-print-samples \
-    --save-dir checkpoints/transformer \
-    --no-epoch-checkpoints \
-    --best-checkpoint-metric bleu --maximize-best-checkpoint-metric
-
-# Evaluate
-fairseq-generate data-bin/iwslt14.tokenized.de-en \
-    --path checkpoints/transformer/checkpoint_best.pt \
-    --batch-size 128 --beam 5 --remove-bpe
-```
-
-## Model Results
-
-| Model       | GPUs       | QPS     | Train Epochs | Bleu  |
-|-------------|------------|---------|--------------|-------|--|
-| Transformer | BI-V100 x8 | 3204.78 | 100          | 35.07 |
-
-## References
-
-- [Fairseq](https://github.com/facebookresearch/fairseq/tree/v0.10.2)
+# Transformer
+
+## Model Description
+
+The Transformer model revolutionized natural language processing with its attention-based architecture, eliminating the
+need for recurrent connections. It employs self-attention mechanisms to process input sequences in parallel, capturing
+long-range dependencies more effectively than previous models. Transformers excel in tasks like translation, text
+generation, and summarization by dynamically weighting the importance of different words in a sequence. Their parallel
+processing capability enables faster training and better scalability, making them the foundation for state-of-the-art
+language models like BERT and GPT.
+
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
+## Model Preparation
+
+### Prepare Resources
+
+```bash
+# Go to "toolbox/Fairseq" directory in root path
+cd ../../../../toolbox/Fairseq/
+
+cd fairseq/examples/translation/
+bash prepare-iwslt14.sh
+cd ../..
+
+TEXT=examples/translation/iwslt14.tokenized.de-en
+fairseq-preprocess --source-lang de --target-lang en \
+    --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \
+    --destdir data-bin/iwslt14.tokenized.de-en \
+    --workers 20
+```
+
+### Install Dependencies
+
+Transformer model is using Fairseq toolbox. Before you run this model, you need to setup Fairseq first.
+
+```bash
+bash install_toolbox_fairseq.sh
+```
+
+## Model Training
+
+```bash
+# Train
+mkdir -p checkpoints/transformer
+fairseq-train data-bin/iwslt14.tokenized.de-en \
+    --arch transformer_iwslt_de_en --share-decoder-input-output-embed \
+    --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \
+    --lr 5e-4 --lr-scheduler inverse_sqrt --warmup-updates 4000 \
+    --dropout 0.3 --weight-decay 0.0001 \
+    --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
+    --max-tokens 4096 \
+    --max-epoch 100 \
+    --eval-bleu \
+    --eval-bleu-args '{"beam": 5, "max_len_a": 1.2, "max_len_b": 10}' \
+    --eval-bleu-detok moses \
+    --eval-bleu-remove-bpe \
+    --eval-bleu-print-samples \
+    --save-dir checkpoints/transformer \
+    --no-epoch-checkpoints \
+    --best-checkpoint-metric bleu --maximize-best-checkpoint-metric
+
+# Evaluate
+fairseq-generate data-bin/iwslt14.tokenized.de-en \
+    --path checkpoints/transformer/checkpoint_best.pt \
+    --batch-size 128 --beam 5 --remove-bpe
+```
+
+## Model Results
+
+| Model       | GPUs       | QPS     | Train Epochs | Bleu  |
+|-------------|------------|---------|--------------|-------|--|
+| Transformer | BI-V100 x8 | 3204.78 | 100          | 35.07 |
+
+## References
+
+- [Fairseq](https://github.com/facebookresearch/fairseq/tree/v0.10.2)
diff --git a/others/graph_machine_learning/graph_wavenet/pytorch/README.md b/others/graph_machine_learning/graph_wavenet/pytorch/README.md
index 58085b498..4194b5237 100644
--- a/others/graph_machine_learning/graph_wavenet/pytorch/README.md
+++ b/others/graph_machine_learning/graph_wavenet/pytorch/README.md
@@ -9,6 +9,12 @@ effectively handles complex, large-scale datasets, demonstrating superior perfor
 WaveNet's innovative approach to modeling both spatial and temporal dependencies makes it a powerful tool for analyzing
 and predicting patterns in dynamic, interconnected systems.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/kolmogorov_arnold_networks/kan/pytorch/README.md b/others/kolmogorov_arnold_networks/kan/pytorch/README.md
index e5b38e962..a8d2b5514 100644
--- a/others/kolmogorov_arnold_networks/kan/pytorch/README.md
+++ b/others/kolmogorov_arnold_networks/kan/pytorch/README.md
@@ -8,6 +8,12 @@ Kolmogorov-Arnold representation theorem. KANs and MLPs are dual: KANs have acti
 have activation functions on nodes. This simple change makes KANs better (sometimes much better!) than MLPs in terms of
 both model accuracy and interpretability.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.1.1     |  24.12  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/others/model_pruning/network-slimming/pytorch/README.md b/others/model_pruning/network-slimming/pytorch/README.md
index 460b0cf14..61c009e83 100755
--- a/others/model_pruning/network-slimming/pytorch/README.md
+++ b/others/model_pruning/network-slimming/pytorch/README.md
@@ -9,6 +9,12 @@ computational costs without sacrificing accuracy. Network Slimming is architectu
 VGG, ResNet, and DenseNet. It's particularly useful for deploying deep learning models on resource-constrained devices,
 offering a balance between model efficiency and predictive performance.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.0.0     |  23.06  |
+
 ## Model Preparation
 
 ### Install Dependencies
diff --git a/others/recommendation_systems/deepfm/paddlepaddle/README.md b/others/recommendation_systems/deepfm/paddlepaddle/README.md
index 39f79e812..de0173daf 100644
--- a/others/recommendation_systems/deepfm/paddlepaddle/README.md
+++ b/others/recommendation_systems/deepfm/paddlepaddle/README.md
@@ -8,6 +8,12 @@ The model is end-to-end trainable and excels in tasks like click-through rate (C
 recommendations. By integrating both FM and DNN, DeepFM efficiently handles sparse data, offering better performance
 compared to traditional methods, especially in large-scale applications such as advertising and product recommendations.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/dlrm/paddlepaddle/README.md b/others/recommendation_systems/dlrm/paddlepaddle/README.md
index bc2bf436d..bb51d7923 100644
--- a/others/recommendation_systems/dlrm/paddlepaddle/README.md
+++ b/others/recommendation_systems/dlrm/paddlepaddle/README.md
@@ -9,6 +9,12 @@ fully-connected layers for numerical features. Its specialized parallelization s
 embedding tables and data parallelism for dense layers, optimizing memory usage and computational efficiency. DLRM
 serves as a benchmark for recommendation system development and performance evaluation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/dlrm/pytorch/README.md b/others/recommendation_systems/dlrm/pytorch/README.md
index 293e9a881..1c0c4ee42 100644
--- a/others/recommendation_systems/dlrm/pytorch/README.md
+++ b/others/recommendation_systems/dlrm/pytorch/README.md
@@ -9,6 +9,12 @@ fully-connected layers for numerical features. Its specialized parallelization s
 embedding tables and data parallelism for dense layers, optimizing memory usage and computational efficiency. DLRM
 serves as a benchmark for recommendation system development and performance evaluation.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/ffm/paddlepaddle/README.md b/others/recommendation_systems/ffm/paddlepaddle/README.md
index a4139f3f6..870881755 100644
--- a/others/recommendation_systems/ffm/paddlepaddle/README.md
+++ b/others/recommendation_systems/ffm/paddlepaddle/README.md
@@ -7,6 +7,12 @@ features of the same field are one-hot separately, so in FFM, each one-dimension
 each field of the other features, which is not only related to the feature, but also to the field. By introducing the
 concept of field, FFM attributes features of the same nature to the same field.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/ncf/pytorch/README.md b/others/recommendation_systems/ncf/pytorch/README.md
index 2461c2f2a..347258681 100644
--- a/others/recommendation_systems/ncf/pytorch/README.md
+++ b/others/recommendation_systems/ncf/pytorch/README.md
@@ -9,6 +9,12 @@ performance. It significantly improves recommendation accuracy by leveraging dee
 particularly effective for collaborative filtering tasks, demonstrating superior results on real-world datasets compared
 to traditional methods. NCF's architecture makes it a powerful tool for personalized recommendation systems.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.2.0     |  22.09  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/wide_deep/paddlepaddle/README.md b/others/recommendation_systems/wide_deep/paddlepaddle/README.md
index 36cf8d967..2ce44043a 100644
--- a/others/recommendation_systems/wide_deep/paddlepaddle/README.md
+++ b/others/recommendation_systems/wide_deep/paddlepaddle/README.md
@@ -8,6 +8,12 @@ for learning complex patterns. This architecture effectively balances precise me
 ability to generalize to unseen combinations. Wide&Deep has proven particularly effective in large-scale recommendation
 systems, offering improved performance in tasks like app recommendation while maintaining computational efficiency.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 2.3.0     |  22.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/others/recommendation_systems/xdeepfm/paddlepaddle/README.md b/others/recommendation_systems/xdeepfm/paddlepaddle/README.md
index 642c1b701..536f79b95 100644
--- a/others/recommendation_systems/xdeepfm/paddlepaddle/README.md
+++ b/others/recommendation_systems/xdeepfm/paddlepaddle/README.md
@@ -9,6 +9,12 @@ networks, enabling both explicit and implicit feature learning. This architectur
 engineering while improving recommendation accuracy. Particularly effective for sparse data, xDeepFM excels in tasks
 like click-through rate prediction, offering enhanced performance in large-scale recommendation scenarios.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V100 | 3.1.0     |  23.12  |
+
 ## Model Preparation
 
 ### Prepare Resources
diff --git a/reinforcement_learning/q-learning-networks/dqn/paddlepaddle/README.md b/reinforcement_learning/q-learning-networks/dqn/paddlepaddle/README.md
index f8b45af44..d5953d124 100644
--- a/reinforcement_learning/q-learning-networks/dqn/paddlepaddle/README.md
+++ b/reinforcement_learning/q-learning-networks/dqn/paddlepaddle/README.md
@@ -8,6 +8,12 @@ DQN introduces experience replay and target network stabilization to enable stab
 revolutionized AI capabilities in complex environments, achieving human-level performance in Atari games and forming the
 basis for advanced decision-making systems in robotics and game AI.
 
+## Supported Environments
+
+| GPU    | [IXUCA SDK](https://gitee.com/deep-spark/deepspark#%E5%A4%A9%E6%95%B0%E6%99%BA%E7%AE%97%E8%BD%AF%E4%BB%B6%E6%A0%88-ixuca) | Release |
+|--------|-----------|---------|
+| BI-V150 | 4.2.0     |  25.03  |
+
 ## Model Preparation
 
 ### Install Dependencies
-- 
Gitee