From 005fe1c0eb6c176593c1a911b23dc2fe5ea8f6b4 Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:20:01 +0800 Subject: [PATCH 1/8] add 25.03 igie models to model list --- README.md | 144 +++++++++++++++++- models/cv/classification/vgg11/igie/README.md | 2 +- .../wide_resnet101/igie/README.md | 2 +- models/cv/detection/sabl/igie/README.md | 1 + .../pose_estimation/hrnetpose/igie/README.md | 5 +- 5 files changed, 145 insertions(+), 9 deletions(-) diff --git a/README.md b/README.md index a92b22a3..f129b927 100644 --- a/README.md +++ b/README.md @@ -90,7 +90,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 CSPResNet50 FP16 - - + Supported Supported @@ -197,6 +197,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + EfficientNet-B4 + FP16 + Supported + - + + + INT8 + - + - + EfficientNetV2 FP16 @@ -274,6 +285,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - Supported + + Mixer_B + FP16 + Supported + - + + + INT8 + - + - + MNASNet0_5 FP16 @@ -285,6 +307,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + MNASNet0_75 + FP16 + Supported + - + + + INT8 + - + - + MobileNetV2 FP16 @@ -329,6 +362,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + RegNet_x_16gf + FP16 + Supported + - + + + INT8 + - + - + RegNet_x_1_6gf FP16 @@ -472,6 +516,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + ResNeXt101_32x8d + FP16 + Supported + - + + + INT8 + - + - + SEResNet50 FP16 @@ -528,26 +583,48 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - SqueezeNet 1.0 + ShuffleNetV2_x2_0 FP16 + Supported - - Supported INT8 - - Supported + - + + + SqueezeNet 1.0 + FP16 + Supported + Supported + + + INT8 + - + Supported SqueezeNet 1.1 FP16 - - Supported + Supported + + + INT8 + - + Supported + + + SVT Base + FP16 + Supported + - INT8 - - Supported + - Swin Transformer @@ -571,6 +648,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + VGG11 + FP16 + Supported + - + + + INT8 + - + - + VGG16 FP16 @@ -593,6 +681,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 Supported Supported + + Wide ResNet101 + FP16 + Supported + - + + + INT8 + - + - + ### Detection @@ -725,6 +824,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + SABL + FP16 + Supported + - + + + INT8 + - + - + YOLOv3 FP16 @@ -824,6 +934,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - + + YOLOv11 + FP16 + Supported + - + + + INT8 + - + - + YOLOX FP16 @@ -902,6 +1023,17 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 IGIE IxRT + + HRNetPose + FP16 + Supported + - + + + INT8 + - + - + Lightweight OpenPose FP16 diff --git a/models/cv/classification/vgg11/igie/README.md b/models/cv/classification/vgg11/igie/README.md index 0206d951..522ff3e7 100644 --- a/models/cv/classification/vgg11/igie/README.md +++ b/models/cv/classification/vgg11/igie/README.md @@ -43,4 +43,4 @@ bash scripts/infer_vgg11_fp16_performance.sh Model |BatchSize |Precision |FPS |Top-1(%) |Top-5(%) --------|-----------|----------|----------|----------|-------- -VGG11 | 32 | FP16 | 3872.86 | 69.03 | 88.6 \ No newline at end of file +VGG11 | 32 | FP16 | 3872.86 | 69.03 | 88.6 diff --git a/models/cv/classification/wide_resnet101/igie/README.md b/models/cv/classification/wide_resnet101/igie/README.md index a72eeb1c..93a5a3b8 100644 --- a/models/cv/classification/wide_resnet101/igie/README.md +++ b/models/cv/classification/wide_resnet101/igie/README.md @@ -43,4 +43,4 @@ bash scripts/infer_wide_resnet101_fp16_performance.sh | Model | BatchSize | Precision | FPS | Top-1(%) | Top-5(%) | | -------------- | --------- | --------- | -------- | -------- | -------- | -| Wide ResNet101 | 32 | FP16 | 1339.037 | 78.459 | 94.052 | \ No newline at end of file +| Wide ResNet101 | 32 | FP16 | 1339.037 | 78.459 | 94.052 | diff --git a/models/cv/detection/sabl/igie/README.md b/models/cv/detection/sabl/igie/README.md index 975e72da..28d0242e 100644 --- a/models/cv/detection/sabl/igie/README.md +++ b/models/cv/detection/sabl/igie/README.md @@ -27,6 +27,7 @@ Dataset: to download the valida ```bash wget https://download.openmmlab.com/mmdetection/v2.0/sabl/sabl_retinanet_r50_fpn_1x_coco/sabl_retinanet_r50_fpn_1x_coco-6c54fd4f.pth ``` + ### Model Conversion ```bash diff --git a/models/cv/pose_estimation/hrnetpose/igie/README.md b/models/cv/pose_estimation/hrnetpose/igie/README.md index 1785d1f7..c4f0758f 100644 --- a/models/cv/pose_estimation/hrnetpose/igie/README.md +++ b/models/cv/pose_estimation/hrnetpose/igie/README.md @@ -5,6 +5,7 @@ HRNetPose (High-Resolution Network for Pose Estimation) is a high-performance human pose estimation model introduced in the paper "Deep High-Resolution Representation Learning for Human Pose Estimation". It is designed to address the limitations of traditional methods by maintaining high-resolution feature representations throughout the network, enabling more accurate detection of human keypoints. ## Setup + ### Install ```bash @@ -18,6 +19,7 @@ pip3 install -r requirements.txt ``` ### Download + Pretrained model: Dataset: to download the validation dataset. @@ -27,6 +29,7 @@ wget https://download.openmmlab.com/mmpose/top_down/hrnet/hrnet_w32_coco_256x192 ``` ### Model Conversion + ```bash # export onnx model python3 export.py --weight hrnet_w32_coco_256x192-c78dce93_20200708.pth --cfg td-hm_hrnet-w32_8xb64-210e_coco-256x192.py --input 1,3,256,192 --output hrnetpose.onnx @@ -58,4 +61,4 @@ bash scripts/infer_hrnetpose_fp16_performance.sh ## Reference -mmpose: \ No newline at end of file +mmpose: -- Gitee From 6bab4f9f68a3b5099b616df09f961a28be484c09 Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:36:08 +0800 Subject: [PATCH 2/8] move vision language understanding models Signed-off-by: mingjiang.li --- .../Intern_VL/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../Intern_VL/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../LLava/vllm/README.md | 0 .../LLava/vllm/offline_inference_vision_language.py | 0 .../LLava/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../chameleon-7b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../chameleon-7b/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../fuyu-8b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../fuyu-8b/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../llava_next_video-7b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../llava_next_video-7b/vllm/utils.py | 0 .../vllm/video-eample-data/sample_demo_1.mp4 | Bin 20 files changed, 0 insertions(+), 0 deletions(-) rename models/{ => multimodal}/vision-language-understanding/Intern_VL/vllm/README.md (100%) rename models/{ => multimodal}/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py (100%) rename models/{ => multimodal}/vision-language-understanding/Intern_VL/vllm/utils.py (100%) rename models/{ => multimodal}/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/{ => multimodal}/vision-language-understanding/LLava/vllm/README.md (100%) rename models/{ => multimodal}/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py (100%) rename models/{ => multimodal}/vision-language-understanding/LLava/vllm/utils.py (100%) rename models/{ => multimodal}/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/{ => multimodal}/vision-language-understanding/chameleon-7b/vllm/README.md (100%) rename models/{ => multimodal}/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py (100%) rename models/{ => multimodal}/vision-language-understanding/chameleon-7b/vllm/utils.py (100%) rename models/{ => multimodal}/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/{ => multimodal}/vision-language-understanding/fuyu-8b/vllm/README.md (100%) rename models/{ => multimodal}/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py (100%) rename models/{ => multimodal}/vision-language-understanding/fuyu-8b/vllm/utils.py (100%) rename models/{ => multimodal}/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/{ => multimodal}/vision-language-understanding/llava_next_video-7b/vllm/README.md (100%) rename models/{ => multimodal}/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py (100%) rename models/{ => multimodal}/vision-language-understanding/llava_next_video-7b/vllm/utils.py (100%) rename models/{ => multimodal}/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 (100%) diff --git a/models/vision-language-understanding/Intern_VL/vllm/README.md b/models/multimodal/vision-language-understanding/Intern_VL/vllm/README.md similarity index 100% rename from models/vision-language-understanding/Intern_VL/vllm/README.md rename to models/multimodal/vision-language-understanding/Intern_VL/vllm/README.md diff --git a/models/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py similarity index 100% rename from models/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py diff --git a/models/vision-language-understanding/Intern_VL/vllm/utils.py b/models/multimodal/vision-language-understanding/Intern_VL/vllm/utils.py similarity index 100% rename from models/vision-language-understanding/Intern_VL/vllm/utils.py rename to models/multimodal/vision-language-understanding/Intern_VL/vllm/utils.py diff --git a/models/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/vision-language-understanding/LLava/vllm/README.md b/models/multimodal/vision-language-understanding/LLava/vllm/README.md similarity index 100% rename from models/vision-language-understanding/LLava/vllm/README.md rename to models/multimodal/vision-language-understanding/LLava/vllm/README.md diff --git a/models/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py similarity index 100% rename from models/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py diff --git a/models/vision-language-understanding/LLava/vllm/utils.py b/models/multimodal/vision-language-understanding/LLava/vllm/utils.py similarity index 100% rename from models/vision-language-understanding/LLava/vllm/utils.py rename to models/multimodal/vision-language-understanding/LLava/vllm/utils.py diff --git a/models/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/vision-language-understanding/chameleon-7b/vllm/README.md b/models/multimodal/vision-language-understanding/chameleon-7b/vllm/README.md similarity index 100% rename from models/vision-language-understanding/chameleon-7b/vllm/README.md rename to models/multimodal/vision-language-understanding/chameleon-7b/vllm/README.md diff --git a/models/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py diff --git a/models/vision-language-understanding/chameleon-7b/vllm/utils.py b/models/multimodal/vision-language-understanding/chameleon-7b/vllm/utils.py similarity index 100% rename from models/vision-language-understanding/chameleon-7b/vllm/utils.py rename to models/multimodal/vision-language-understanding/chameleon-7b/vllm/utils.py diff --git a/models/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/vision-language-understanding/fuyu-8b/vllm/README.md b/models/multimodal/vision-language-understanding/fuyu-8b/vllm/README.md similarity index 100% rename from models/vision-language-understanding/fuyu-8b/vllm/README.md rename to models/multimodal/vision-language-understanding/fuyu-8b/vllm/README.md diff --git a/models/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py diff --git a/models/vision-language-understanding/fuyu-8b/vllm/utils.py b/models/multimodal/vision-language-understanding/fuyu-8b/vllm/utils.py similarity index 100% rename from models/vision-language-understanding/fuyu-8b/vllm/utils.py rename to models/multimodal/vision-language-understanding/fuyu-8b/vllm/utils.py diff --git a/models/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/vision-language-understanding/llava_next_video-7b/vllm/README.md b/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/README.md similarity index 100% rename from models/vision-language-understanding/llava_next_video-7b/vllm/README.md rename to models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/README.md diff --git a/models/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py diff --git a/models/vision-language-understanding/llava_next_video-7b/vllm/utils.py b/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/utils.py similarity index 100% rename from models/vision-language-understanding/llava_next_video-7b/vllm/utils.py rename to models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/utils.py diff --git a/models/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 b/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 similarity index 100% rename from models/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 rename to models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 -- Gitee From d88675ccc713b35244239cedd35ae7e32933476f Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:40:29 +0800 Subject: [PATCH 3/8] rename model path of vision language understanding models Signed-off-by: mingjiang.li --- .../{chameleon-7b => chameleon_7b}/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../{chameleon-7b => chameleon_7b}/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../{fuyu-8b => fuyu_8b}/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../{fuyu-8b => fuyu_8b}/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../vllm/utils.py | 0 .../vllm/video-eample-data/sample_demo_1.mp4 | Bin .../{minicpm-v-2 => minicpm_v_2}/vllm/README.md | 0 .../{minicpm-v-2 => minicpm_v_2}/vllm/ci/prepare.sh | 0 .../vllm/minicpmv-2.0-offline.py | 0 15 files changed, 0 insertions(+), 0 deletions(-) rename models/multimodal/vision-language-understanding/{chameleon-7b => chameleon_7b}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{chameleon-7b => chameleon_7b}/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/vision-language-understanding/{chameleon-7b => chameleon_7b}/vllm/utils.py (100%) rename models/multimodal/vision-language-understanding/{chameleon-7b => chameleon_7b}/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/vision-language-understanding/{fuyu-8b => fuyu_8b}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{fuyu-8b => fuyu_8b}/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/vision-language-understanding/{fuyu-8b => fuyu_8b}/vllm/utils.py (100%) rename models/multimodal/vision-language-understanding/{fuyu-8b => fuyu_8b}/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/vision-language-understanding/{llava_next_video-7b => llava_next_video_7b}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{llava_next_video-7b => llava_next_video_7b}/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/vision-language-understanding/{llava_next_video-7b => llava_next_video_7b}/vllm/utils.py (100%) rename models/multimodal/vision-language-understanding/{llava_next_video-7b => llava_next_video_7b}/vllm/video-eample-data/sample_demo_1.mp4 (100%) rename models/multimodal/vision-language-understanding/{minicpm-v-2 => minicpm_v_2}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{minicpm-v-2 => minicpm_v_2}/vllm/ci/prepare.sh (100%) rename models/multimodal/vision-language-understanding/{minicpm-v-2 => minicpm_v_2}/vllm/minicpmv-2.0-offline.py (100%) diff --git a/models/multimodal/vision-language-understanding/chameleon-7b/vllm/README.md b/models/multimodal/vision-language-understanding/chameleon_7b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon-7b/vllm/README.md rename to models/multimodal/vision-language-understanding/chameleon_7b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/chameleon_7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon-7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/chameleon_7b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/chameleon-7b/vllm/utils.py b/models/multimodal/vision-language-understanding/chameleon_7b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon-7b/vllm/utils.py rename to models/multimodal/vision-language-understanding/chameleon_7b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon-7b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/fuyu-8b/vllm/README.md b/models/multimodal/vision-language-understanding/fuyu_8b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu-8b/vllm/README.md rename to models/multimodal/vision-language-understanding/fuyu_8b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/fuyu_8b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu-8b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/fuyu_8b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/fuyu-8b/vllm/utils.py b/models/multimodal/vision-language-understanding/fuyu_8b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu-8b/vllm/utils.py rename to models/multimodal/vision-language-understanding/fuyu_8b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu-8b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/README.md b/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/README.md rename to models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/utils.py b/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/utils.py rename to models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 b/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video-7b/vllm/video-eample-data/sample_demo_1.mp4 rename to models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 diff --git a/models/multimodal/vision-language-understanding/minicpm-v-2/vllm/README.md b/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm-v-2/vllm/README.md rename to models/multimodal/vision-language-understanding/minicpm_v_2/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/minicpm-v-2/vllm/ci/prepare.sh b/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/ci/prepare.sh similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm-v-2/vllm/ci/prepare.sh rename to models/multimodal/vision-language-understanding/minicpm_v_2/vllm/ci/prepare.sh diff --git a/models/multimodal/vision-language-understanding/minicpm-v-2/vllm/minicpmv-2.0-offline.py b/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm-v-2/vllm/minicpmv-2.0-offline.py rename to models/multimodal/vision-language-understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py -- Gitee From 6b20ea4f6807b7a47cd5804eb0813f80bd43d2b0 Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:41:49 +0800 Subject: [PATCH 4/8] rename intern_vl model path Signed-off-by: mingjiang.li --- .../{Intern_VL => intern_vl}/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../{Intern_VL => intern_vl}/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin 4 files changed, 0 insertions(+), 0 deletions(-) rename models/multimodal/vision-language-understanding/{Intern_VL => intern_vl}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{Intern_VL => intern_vl}/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/vision-language-understanding/{Intern_VL => intern_vl}/vllm/utils.py (100%) rename models/multimodal/vision-language-understanding/{Intern_VL => intern_vl}/vllm/vllm_public_assets/cherry_blossom.jpg (100%) diff --git a/models/multimodal/vision-language-understanding/Intern_VL/vllm/README.md b/models/multimodal/vision-language-understanding/intern_vl/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/Intern_VL/vllm/README.md rename to models/multimodal/vision-language-understanding/intern_vl/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/intern_vl/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/Intern_VL/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/intern_vl/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/Intern_VL/vllm/utils.py b/models/multimodal/vision-language-understanding/intern_vl/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/Intern_VL/vllm/utils.py rename to models/multimodal/vision-language-understanding/intern_vl/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/Intern_VL/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg -- Gitee From aa1b82bf1c5697468ccaeea142f7cac37a34408c Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:42:42 +0800 Subject: [PATCH 5/8] rename llava model path Signed-off-by: mingjiang.li --- .../{LLava => llava}/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../{LLava => llava}/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin 4 files changed, 0 insertions(+), 0 deletions(-) rename models/multimodal/vision-language-understanding/{LLava => llava}/vllm/README.md (100%) rename models/multimodal/vision-language-understanding/{LLava => llava}/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/vision-language-understanding/{LLava => llava}/vllm/utils.py (100%) rename models/multimodal/vision-language-understanding/{LLava => llava}/vllm/vllm_public_assets/cherry_blossom.jpg (100%) diff --git a/models/multimodal/vision-language-understanding/LLava/vllm/README.md b/models/multimodal/vision-language-understanding/llava/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/LLava/vllm/README.md rename to models/multimodal/vision-language-understanding/llava/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py b/models/multimodal/vision-language-understanding/llava/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/LLava/vllm/offline_inference_vision_language.py rename to models/multimodal/vision-language-understanding/llava/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/LLava/vllm/utils.py b/models/multimodal/vision-language-understanding/llava/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/LLava/vllm/utils.py rename to models/multimodal/vision-language-understanding/llava/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision-language-understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/LLava/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision-language-understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg -- Gitee From 3fabc4087eeb872ae6215752f2ee128c204027ba Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:43:47 +0800 Subject: [PATCH 6/8] rename vision_language_understanding path Signed-off-by: mingjiang.li --- .../chameleon_7b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../chameleon_7b/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../fuyu_8b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../fuyu_8b/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../intern_vl/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../intern_vl/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../llava/vllm/README.md | 0 .../llava/vllm/offline_inference_vision_language.py | 0 .../llava/vllm/utils.py | 0 .../vllm/vllm_public_assets/cherry_blossom.jpg | Bin .../llava_next_video_7b/vllm/README.md | 0 .../vllm/offline_inference_vision_language.py | 0 .../llava_next_video_7b/vllm/utils.py | 0 .../vllm/video-eample-data/sample_demo_1.mp4 | Bin .../minicpm_v_2/vllm/README.md | 0 .../minicpm_v_2/vllm/ci/prepare.sh | 0 .../minicpm_v_2/vllm/minicpmv-2.0-offline.py | 0 23 files changed, 0 insertions(+), 0 deletions(-) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/chameleon_7b/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/chameleon_7b/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/chameleon_7b/vllm/utils.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/fuyu_8b/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/fuyu_8b/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/fuyu_8b/vllm/utils.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/intern_vl/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/intern_vl/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/intern_vl/vllm/utils.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava/vllm/utils.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava/vllm/vllm_public_assets/cherry_blossom.jpg (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava_next_video_7b/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava_next_video_7b/vllm/offline_inference_vision_language.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava_next_video_7b/vllm/utils.py (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/minicpm_v_2/vllm/README.md (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/minicpm_v_2/vllm/ci/prepare.sh (100%) rename models/multimodal/{vision-language-understanding => vision_language_understanding}/minicpm_v_2/vllm/minicpmv-2.0-offline.py (100%) diff --git a/models/multimodal/vision-language-understanding/chameleon_7b/vllm/README.md b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon_7b/vllm/README.md rename to models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/chameleon_7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon_7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision_language_understanding/chameleon_7b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/chameleon_7b/vllm/utils.py b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon_7b/vllm/utils.py rename to models/multimodal/vision_language_understanding/chameleon_7b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision_language_understanding/chameleon_7b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/fuyu_8b/vllm/README.md b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu_8b/vllm/README.md rename to models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/fuyu_8b/vllm/offline_inference_vision_language.py b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu_8b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision_language_understanding/fuyu_8b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/fuyu_8b/vllm/utils.py b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu_8b/vllm/utils.py rename to models/multimodal/vision_language_understanding/fuyu_8b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision_language_understanding/fuyu_8b/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/intern_vl/vllm/README.md b/models/multimodal/vision_language_understanding/intern_vl/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/intern_vl/vllm/README.md rename to models/multimodal/vision_language_understanding/intern_vl/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/intern_vl/vllm/offline_inference_vision_language.py b/models/multimodal/vision_language_understanding/intern_vl/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/intern_vl/vllm/offline_inference_vision_language.py rename to models/multimodal/vision_language_understanding/intern_vl/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/intern_vl/vllm/utils.py b/models/multimodal/vision_language_understanding/intern_vl/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/intern_vl/vllm/utils.py rename to models/multimodal/vision_language_understanding/intern_vl/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision_language_understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision_language_understanding/intern_vl/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/llava/vllm/README.md b/models/multimodal/vision_language_understanding/llava/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/llava/vllm/README.md rename to models/multimodal/vision_language_understanding/llava/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/llava/vllm/offline_inference_vision_language.py b/models/multimodal/vision_language_understanding/llava/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava/vllm/offline_inference_vision_language.py rename to models/multimodal/vision_language_understanding/llava/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/llava/vllm/utils.py b/models/multimodal/vision_language_understanding/llava/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava/vllm/utils.py rename to models/multimodal/vision_language_understanding/llava/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg b/models/multimodal/vision_language_understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg similarity index 100% rename from models/multimodal/vision-language-understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg rename to models/multimodal/vision_language_understanding/llava/vllm/vllm_public_assets/cherry_blossom.jpg diff --git a/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/README.md b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/README.md rename to models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py rename to models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/offline_inference_vision_language.py diff --git a/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/utils.py b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/utils.py similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/utils.py rename to models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/utils.py diff --git a/models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 similarity index 100% rename from models/multimodal/vision-language-understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 rename to models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/video-eample-data/sample_demo_1.mp4 diff --git a/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/README.md b/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm_v_2/vllm/README.md rename to models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md diff --git a/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/ci/prepare.sh b/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/ci/prepare.sh similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm_v_2/vllm/ci/prepare.sh rename to models/multimodal/vision_language_understanding/minicpm_v_2/vllm/ci/prepare.sh diff --git a/models/multimodal/vision-language-understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py b/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py similarity index 100% rename from models/multimodal/vision-language-understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py rename to models/multimodal/vision_language_understanding/minicpm_v_2/vllm/minicpmv-2.0-offline.py -- Gitee From 32f0fd76914515292387079310a1d063117407c3 Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 17:49:25 +0800 Subject: [PATCH 7/8] add ixrt small models to model list Signed-off-by: mingjiang.li --- README.md | 18 +++++++++--------- .../cv/classification/deit_tiny/ixrt/README.md | 2 +- .../minicpm_v_2/vllm/README.md | 4 ++-- 3 files changed, 12 insertions(+), 12 deletions(-) diff --git a/README.md b/README.md index f129b927..e02ec70e 100644 --- a/README.md +++ b/README.md @@ -58,7 +58,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 ConvNeXt-Base FP16 Supported - - + Supported INT8 @@ -102,7 +102,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 DeiT-tiny FP16 Supported - - + Supported INT8 @@ -146,7 +146,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 DenseNet201 FP16 Supported - - + Supported INT8 @@ -223,7 +223,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 EfficientNetv2_rw_t FP16 Supported - - + Supported INT8 @@ -751,7 +751,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 FoveaBox FP16 Supported - - + Supported INT8 @@ -762,7 +762,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 FSAF FP16 Supported - - + Supported INT8 @@ -773,7 +773,7 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 HRNet FP16 Supported - - + Supported INT8 @@ -1311,8 +1311,8 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 - - MiniCPM-V-2 - Supported + MiniCPM V2 + Supported - - diff --git a/models/cv/classification/deit_tiny/ixrt/README.md b/models/cv/classification/deit_tiny/ixrt/README.md index 1ceea7bf..15fb025e 100644 --- a/models/cv/classification/deit_tiny/ixrt/README.md +++ b/models/cv/classification/deit_tiny/ixrt/README.md @@ -70,4 +70,4 @@ bash scripts/infer_deit_tiny_fp16_performance.sh ## Reference -Deit_tiny: \ No newline at end of file +Deit_tiny: diff --git a/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md b/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md index 2dc49881..149f01f1 100644 --- a/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md +++ b/models/multimodal/vision_language_understanding/minicpm_v_2/vllm/README.md @@ -1,8 +1,8 @@ -# MiniCPM-V-2 +# MiniCPM V2 ## Description -MiniCPM-V-2 is a compact and efficient language model designed for various natural language processing (NLP) tasks. Building on its predecessor, MiniCPM-V-1, this model integrates advancements in architecture and optimization techniques, making it suitable for deployment in resource-constrained environments.s +MiniCPM V2 is a compact and efficient language model designed for various natural language processing (NLP) tasks. Building on its predecessor, MiniCPM-V-1, this model integrates advancements in architecture and optimization techniques, making it suitable for deployment in resource-constrained environments.s ## Setup -- Gitee From 6cd8166cade544e9e7c39ecaabcec3b7bedc6309 Mon Sep 17 00:00:00 2001 From: "mingjiang.li" Date: Tue, 11 Feb 2025 18:17:16 +0800 Subject: [PATCH 8/8] move llm and multimodal category to h2 level Signed-off-by: mingjiang.li --- README.md | 222 ++++++++++-------- .../chameleon_7b/vllm/README.md | 4 +- .../fuyu_8b/vllm/README.md | 4 +- .../intern_vl/vllm/README.md | 3 +- .../llava/vllm/README.md | 5 +- .../llava_next_video_7b/vllm/README.md | 4 +- 6 files changed, 131 insertions(+), 111 deletions(-) diff --git a/README.md b/README.md index e02ec70e..c9932f74 100644 --- a/README.md +++ b/README.md @@ -988,29 +988,24 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 Models Precision IGIE - IxRT Kie_layoutXLM FP16 Supported - - INT8 - - - SVTR FP16 Supported - - INT8 - - - @@ -1158,209 +1153,236 @@ DeepSparkInference将按季度进行版本更新，后续会逐步丰富模型 -## NLP - -### Language Modelling +## LLM (Large Language Model) - - - + + + - - + + + - - + + - - + + - - + + + - - - - - - - + + + - - - - - + + + + - - - + + + + - - + + - + - - + + + - - + + + - - + + - - + + + - - + + - - + + + - - + +
ModelsPrecisionIGIEIxRTvLLMTensorRT-LLMTGI
ALBERTFP16Baichuan2-7BSupported- -Supported
INT8ChatGLM-3-6BSupported - -
BERT Base NERFP16ChatGLM-3-6B-32KSupported - -
INT8SupportedLlama2-7BSupportedSupported -
BERT Base SQuADFP16SupportedSupported
INT8Llama2-13B-Supported -Supported
BERT Large SQuADFP16SupportedSupportedLlama2-70B-Supported-
INT8SupportedSupportedLlama3-70BSupported--
DeBERTaFP16Qwen-7BSupported -Supported-
INT8-Qwen1.5-7BSupported -Supported
RoBERTaFP16Qwen1.5-14BSupported- -Supported
INT8Qwen1.5-32B ChatSupported - -
RoFormerFP16Qwen1.5-72BSupported- -Supported
INT8Qwen2-7B InstructSupported - -
VideoBERTFP16Qwen2-72B InstructSupported- -Supported
INT8StableLM2-1.6BSupported - -
-### Large Language Model +## Multimodal - - + + + + + + + + + + + + + + + + + + + + + + + + +
Models vLLMTensorRT-LLMTGI
Chameleon-7BSupported
Fuyu-8BSupported
InternVL2-4BSupported
LLaVASupported
LLaVA-Next-Video-7BSupported
MiniCPM V2Supported
+ +## NLP + +### Language Modelling + + - - - + + + + + + + + + - - + - - + + - - - + + - - - - + + + + - - - + + - - - - + + + + - - - - + + + - - + + - + - - + + - - - - + + + - - + - - - + + + - - + - - - + + + - - + diff --git a/models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md index bebd7c79..568dbd70 100755 --- a/models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md +++ b/models/multimodal/vision_language_understanding/chameleon_7b/vllm/README.md @@ -2,7 +2,7 @@ ## Description -Chameleon, an AI system that mitigates these limitations by augmenting LLMs with plug-and-play modules for compositional reasoning. Chameleon synthesizes programs by composing various tools (e.g., LLMs, off-the-shelf vision models, web search engines, Python functions, and heuristic-based modules) for accomplishing complex reasoning tasks. At the heart of Chameleon is an LLM-based planner that assembles a sequence of tools to execute to generate the final response. We showcase the effectiveness of Chameleon on two multi-modal knowledge-intensive reasoning tasks: ScienceQA and TabMWP. Chameleon, powered by GPT-4, achieves an 86.54% overall accuracy on ScienceQA, improving the best published few-shot result by 11.37%. On TabMWP, GPT-4-powered Chameleon improves the accuracy by 17.0%, lifting the state of the art to 98.78%. Our analysis also shows that the GPT-4-powered planner exhibits more consistent and rational tool selection via inferring potential constraints from instructions, compared to a ChatGPT-powered planner. +Chameleon, an AI system that mitigates these limitations by augmenting LLMs with plug-and-play modules for compositional reasoning. Chameleon synthesizes programs by composing various tools (e.g., LLMs, off-the-shelf vision models, web search engines, Python functions, and heuristic-based modules) for accomplishing complex reasoning tasks. At the heart of Chameleon is an LLM-based planner that assembles a sequence of tools to execute to generate the final response. We showcase the effectiveness of Chameleon on two multi-modal knowledge-intensive reasoning tasks: ScienceQA and TabMWP. Chameleon, powered by GPT-4, achieves an 86.54% overall accuracy on ScienceQA, improving the best published few-shot result by 11.37%. On TabMWP, GPT-4-powered Chameleon improves the accuracy by 17.0%, lifting the state of the art to 98.78%. Our analysis also shows that the GPT-4-powered planner exhibits more consistent and rational tool selection via inferring potential constraints from instructions, compared to a ChatGPT-powered planner. ## Setup @@ -32,4 +32,4 @@ mkdir data ```bash export VLLM_ASSETS_CACHE=../vllm/ python3 offline_inference_vision_language.py --model ./data/chameleon-7b --max-tokens 256 -tp 2 --trust-remote-code --temperature 0.0 -``` \ No newline at end of file +``` diff --git a/models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md index 7bc5d2cc..96d5acc3 100755 --- a/models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md +++ b/models/multimodal/vision_language_understanding/fuyu_8b/vllm/README.md @@ -1,4 +1,4 @@ -# FuyuForCausalLM +# Fuyu-8B ## Description @@ -34,4 +34,4 @@ mkdir data ```bash export VLLM_ASSETS_CACHE=../vllm/ python3 offline_inference_vision_language.py --model ./data/fuyu-8b --max-tokens 256 -tp 2 --trust-remote-code --temperature 0.0 -``` \ No newline at end of file +``` diff --git a/models/multimodal/vision_language_understanding/intern_vl/vllm/README.md b/models/multimodal/vision_language_understanding/intern_vl/vllm/README.md index 0b09f06e..cafe64f9 100644 --- a/models/multimodal/vision_language_understanding/intern_vl/vllm/README.md +++ b/models/multimodal/vision_language_understanding/intern_vl/vllm/README.md @@ -6,7 +6,7 @@ InternVL2-4B is a large-scale multimodal model developed by WeTab AI, designed t ## Setup -### Instal +### Install In order to run the model smoothly, you need to get the sdk from [resource center](https://support.iluvatar.com/#/ProductLine?id=2) of Iluvatar CoreX official website. @@ -22,6 +22,7 @@ pip3 install vllm pip3 install triton pip3 install ixformer ``` + ### Download -Model: diff --git a/models/multimodal/vision_language_understanding/llava/vllm/README.md b/models/multimodal/vision_language_understanding/llava/vllm/README.md index 1b805f41..bbc251d2 100644 --- a/models/multimodal/vision_language_understanding/llava/vllm/README.md +++ b/models/multimodal/vision_language_understanding/llava/vllm/README.md @@ -4,7 +4,6 @@ LLaVA is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture.The LLaVA-NeXT model was proposed in LLaVA-NeXT: Improved reasoning, OCR, and world knowledge by Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, Yong Jae Lee. LLaVa-NeXT (also called LLaVa-1.6) improves upon LLaVa-1.5 by increasing the input image resolution and training on an improved visual instruction tuning dataset to improve OCR and common sense reasoning. - ## Setup ### Install @@ -24,7 +23,6 @@ pip3 install transformers -llava-v1.6-vicuna-7b-hf: - ```bash # Download model from the website and make sure the model's path is "data/llava" mkdir data @@ -39,11 +37,10 @@ export PATH=/usr/local/corex/bin:${PATH} export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64 ``` - ### Inference llava-1.6 ```bash export VLLM_ASSETS_CACHE=../vllm/ export CUDA_VISIBLE_DEVICES=0,1,2,3 python3 offline_inference_vision_language.py --model /path/to/model --max-tokens 256 -tp 4 --trust-remote-code --temperature 0.0 --model-type llava-next --max-model-len 4096 -``` \ No newline at end of file +``` diff --git a/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md index a50af3a2..bf4b2683 100755 --- a/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md +++ b/models/multimodal/vision_language_understanding/llava_next_video_7b/vllm/README.md @@ -1,4 +1,4 @@ -# LLaVA-Next-Video +# LLaVA-Next-Video-7B ## Description @@ -32,4 +32,4 @@ mkdir data ```bash export VLLM_ASSETS_CACHE=../vllm/ python3 offline_inference_vision_language.py --model ./data/LLaVA-NeXT-Video-7B-hf --max-tokens 256 -tp 4 --trust-remote-code --temperature 0.0 --model-type llava-next-video --modality video --dtype bfloat16 -``` \ No newline at end of file +``` -- Gitee
Baichuan2-7BSupported-ModelsPrecisionIGIEIxRT
ALBERTFP16 -Supported
ChatGLM-3-6BSupportedINT8 - -
ChatGLM-3-6B-32KSupportedBERT Base NERFP16 - -
Llama2-7BSupportedSupportedINT8Supported -
Llama2-13B-Supported-BERT Base SQuADFP16SupportedSupported
Llama2-70B-SupportedINT8 -Supported
Llama3-70BSupported--BERT Large SQuADFP16SupportedSupported
MiniCPM V2Supported--INT8SupportedSupported
Qwen-7BSupportedDeBERTaFP16 --Supported
Qwen1.5-7BSupportedINT8- -Supported
Qwen1.5-14BSupported-RoBERTaFP16 -Supported
Qwen1.5-32B ChatSupportedINT8 - -
Qwen1.5-72BSupported-RoFormerFP16 -Supported
Qwen2-7B InstructSupportedINT8 - -
Qwen2-72B InstructSupported-VideoBERTFP16 -Supported
StableLM2-1.6BSupportedINT8 - -