Update README.md
Browse files
README.md
CHANGED
|
@@ -76,6 +76,26 @@ Since there are still many different metrics that need to be tested, **we chose
|
|
| 76 |
</tr>
|
| 77 |
</table>
|
| 78 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
We are still working on more detailed benchmarks.
|
| 80 |
|
| 81 |
## Examples
|
|
|
|
| 76 |
</tr>
|
| 77 |
</table>
|
| 78 |
|
| 79 |
+
The benchmark result in [MTVQA](https://github.com/bytedance/MTVQA/tree/main)
|
| 80 |
+
|
| 81 |
+
| Models | Open-Source | Vietnamese Score |
|
| 82 |
+
|:----------------------------------:|:-------------:|:------------------:|
|
| 83 |
+
| Qwen2-VL 72B (Top1) | ✗ | 41.6 |
|
| 84 |
+
| GPT-4o (Top2) | ✗ | 34.2 |
|
| 85 |
+
| **Vintern-1B-V2** (Top3) | ✓ | **31.7** |
|
| 86 |
+
| Qwen2-VL 7B | ✓ | 30.0 |
|
| 87 |
+
| Claude3 Opus | ✗ | 29.1 |
|
| 88 |
+
| GPT-4o mini | ✗ | 29.1 |
|
| 89 |
+
| GPT-4V | ✗ | 28.9 |
|
| 90 |
+
| Vintern-1B-V3 | ✓ | 28.7 |
|
| 91 |
+
| Gemini Ultra | ✗ | 28.6 |
|
| 92 |
+
| InternVL2 76B | ✓ | 26.9 |
|
| 93 |
+
| QwenVL Max | ✗ | 23.5 |
|
| 94 |
+
| Claude3 Sonnet | ✗ | 20.8 |
|
| 95 |
+
| QwenVL Plus | ✗ | 18.1 |
|
| 96 |
+
| MiniCPM-V2.5 | ✓ | 15.3 |
|
| 97 |
+
| InternVL-V1.5 | ✗ | 12.4 |
|
| 98 |
+
|
| 99 |
We are still working on more detailed benchmarks.
|
| 100 |
|
| 101 |
## Examples
|