Hugging Face's logo

OpenGVLab
/
InternVL2-Llama3-76B

Image-Text-to-Text
Transformers
Safetensors
multilingual
internvl_chat
feature-extraction
internvl
custom_code
conversational
Model card Files
xet
Community
12
czczup commited on
Commit
d067cc5
·
verified ·
1 Parent(s): c85e698

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +18 -18
README.md CHANGED
@@ -68,15 +68,15 @@ InternVL 2.0 is a multimodal large language model series, featuring models of va
68
 
69
  ### Video Benchmarks
70
 
71
- | Benchmark | GPT-4V | VILA-1.5 | LLaVA-NeXT-Video | InternVL2-40B | InternVL2-Llama3-76B |
72
- | :-------------------------: | :----: | :------: | :--------------: | :-----------: | :------------------: |
73
- | Model Size | - | 34B | 34B | 40B | 76B |
74
- | | | | | | |
75
- | MVBench | - | - | - | 72.5 | 69.6 |
76
- | MMBench-Video<sub>8f</sub> | 1.53 | - | - | 1.32 | 1.37 |
77
- | MMBench-Video<sub>16f</sub> | 1.68 | - | - | 1.45 | 1.52 |
78
- | Video-MME<br>wo subs | 59.9 | 59.0 | 52.0 | TODO | TODO |
79
- | Video-MME<br>w/ subs | 63.3 | 59.4 | 54.9 | TODO | TODO |
80
 
81
  - We evaluate our models on MVBench by extracting 16 frames from each video, and each frame was resized to a 448x448 image.
82
 
@@ -488,15 +488,15 @@ InternVL 2.0 是一个多模态大语言模型系列，包含各种规模的模
488
 
489
  ### 视频相关评测
490
 
491
- | 评测数据集 | GPT-4V | VILA-1.5 | LLaVA-NeXT-Video | InternVL2-40B | InternVL2-Llama3-76B |
492
- | :-------------------------: | :----: | :------: | :--------------: | :-----------: | :------------------: |
493
- | 模型大小 | - | 34B | 34B | 40B | 76B |
494
- | | | | | | |
495
- | MVBench | - | - | - | 72.5 | 69.6 |
496
- | MMBench-Video<sub>8f</sub> | 1.53 | - | - | 1.32 | 1.37 |
497
- | MMBench-Video<sub>16f</sub> | 1.68 | - | - | 1.45 | 1.52 |
498
- | Video-MME<br>wo subs | 59.9 | 59.0 | 52.0 | TODO | TODO |
499
- | Video-MME<br>w/ subs | 63.3 | 59.4 | 54.9 | TODO | TODO |
500
 
501
  - 我们通过从每个视频中提取16帧来评估我们的模型在MVBench上的性能，每个视频帧被调整为448x448的图像。
502
 
 
68
 
69
  ### Video Benchmarks
70
 
71
+ | Benchmark | GPT-4o | GPT-4V | Gemini-Pro-1.5 | InternVL2-40B | InternVL2-Llama3-76B |
72
+ | :-------------------------: | :----: | :----: | :------------: | :-----------: | :------------------: |
73
+ | Model Size | - | - | - | 40B | 76B |
74
+ | | | | | | |
75
+ | MVBench | - | - | - | 72.5 | 69.6 |
76
+ | MMBench-Video<sub>8f</sub> | 1.62 | 1.53 | 1.30 | 1.32 | 1.37 |
77
+ | MMBench-Video<sub>16f</sub> | 1.86 | 1.68 | 1.60 | 1.45 | 1.52 |
78
+ | Video-MME<br>wo subs | 71.9 | 59.9 | 75.0 | TODO | TODO |
79
+ | Video-MME<br>w/ subs | 77.2 | 63.3 | 81.3 | TODO | TODO |
80
 
81
  - We evaluate our models on MVBench by extracting 16 frames from each video, and each frame was resized to a 448x448 image.
82
 
 
488
 
489
  ### 视频相关评测
490
 
491
+ | 评测数据集 | GPT-4o | GPT-4V | Gemini-Pro-1.5 | InternVL2-40B | InternVL2-Llama3-76B |
492
+ | :-------------------------: | :----: | :----: | :------------: | :-----------: | :------------------: |
493
+ | 模型大小 | - | - | - | 40B | 76B |
494
+ | | | | | | |
495
+ | MVBench | - | - | - | 72.5 | 69.6 |
496
+ | MMBench-Video<sub>8f</sub> | 1.62 | 1.53 | 1.30 | 1.32 | 1.37 |
497
+ | MMBench-Video<sub>16f</sub> | 1.86 | 1.68 | 1.60 | 1.45 | 1.52 |
498
+ | Video-MME<br>wo subs | 71.9 | 59.9 | 75.0 | TODO | TODO |
499
+ | Video-MME<br>w/ subs | 77.2 | 63.3 | 81.3 | TODO | TODO |
500
 
501
  - 我们通过从每个视频中提取16帧来评估我们的模型在MVBench上的性能，每个视频帧被调整为448x448的图像。
502