Update README.md
Browse files
README.md
CHANGED
@@ -47,10 +47,10 @@ We are pleased to announce the release of **Ovis2**, our latest advancement in m
|
|
47 |
| Ovis2-34B | aimv2-1B-patch14-448 | Qwen2.5-32B-Instruct | [Huggingface](https://huggingface.co/AIDC-AI/Ovis2-34B) | - |
|
48 |
|
49 |
## Performance
|
50 |
-
.
|
|
|
47 |
| Ovis2-34B | aimv2-1B-patch14-448 | Qwen2.5-32B-Instruct | [Huggingface](https://huggingface.co/AIDC-AI/Ovis2-34B) | - |
|
48 |
|
49 |
## Performance
|
50 |
+

|
51 |
|
52 |
### Image Benchmark
|
53 |
+
| Benchmark | Qwen2.5-VL-7B | InternVL2.5-8B-MPO | MiniCPM-o-2.6 | Ovis1.6-9B | InternVL2.5-4B-MPO | Ovis2-4B | Ovis2-8B |
|
54 |
|:-----------------------------|:---------------:|:--------------------:|:---------------:|:------------:|:--------------------:|:----------:|:----------:|
|
55 |
| MMBench-V1.1<sub>test</sub> | 82.6 | 82.0 | 80.6 | 80.5 | 77.8 | 81.4 | **83.6** |
|
56 |
| MMStar | 64.1 | **65.2** | 63.3 | 62.9 | 61 | 61.9 | 64.6 |
|
|
|
69 |
| MTVQA | 28.5 | 27.2 | 23.1 | 19.2 | 28 | 29.4 | **29.7** |
|
70 |
|
71 |
### Video Benchmark
|
72 |
+
| Benchmark | Qwen2.5-VL-7B | InternVL2.5-8B | LLaVA-OV-7B | InternVL2.5-4B | Ovis2-4B | Ovis2-8B |
|
73 |
+
|:--------------------|:-------------:|:--------------:|:------------------:|:--------------:|:---------:|:-------------:|
|
74 |
+
| VideoMME(wo/w-subs) | 65.1/71.6 | 64.2 / 66.9 | 58.2/61.5 | 62.3 / 63.6 | 64.0/66.3 | **68.0/71.6** |
|
75 |
+
| MVBench | 69.6 | **72.0** | 56.7 | 71.6 | 68.45 | 68.15 |
|
76 |
+
| MLVU(M-Avg/G-Avg) | 70.2/- | 68.9/- | 64.7/- | 68.3/- | 70.8/4.23 | **76.4**/4.25 |
|
77 |
+
| MMBench-Video | 1.79 | 1.68 | - | 1.73 | 1.69 | **1.85** |
|
78 |
+
| TempCompass | **71.7** | - | - | - | 67.02 | 69.28 |
|
79 |
|
80 |
## Usage
|
81 |
Below is a code snippet demonstrating how to run Ovis with various input types. For additional usage instructions, including inference wrapper and Gradio UI, please refer to [Ovis GitHub](https://github.com/AIDC-AI/Ovis?tab=readme-ov-file#inference).
|