Update README.md
Browse files
README.md
CHANGED
|
@@ -97,6 +97,14 @@ Comparison with 30B-70B open-source models:
|
|
| 97 |
| MMLongBench-DOC (Acc) | 42.1 | - | 38.8 | - |
|
| 98 |
</div>
|
| 99 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
|
| 101 |
## 3. Usage
|
| 102 |
|
|
|
|
| 97 |
| MMLongBench-DOC (Acc) | 42.1 | - | 38.8 | - |
|
| 98 |
</div>
|
| 99 |
|
| 100 |
+
Text results, comparison with 30B-level non-thinking VLMs:
|
| 101 |
+
|
| 102 |
+
| Benchmark (Metric) | Kimi-VL-A3B-Thinking-2506 | Qwen2.5-VL-32B | Gemma3-27B-IT |
|
| 103 |
+
|----------------------------|---------------------------|---------------|---------------|
|
| 104 |
+
| MMLU | **82.0** | 78.4 | 76.9 |
|
| 105 |
+
| MMLU-Pro | 68.5 | **68.8** | 67.5 |
|
| 106 |
+
| MATH | **91.8** | 82.2 | 89.0 |
|
| 107 |
+
| GPQA-Diamond | 42.3 | **46.0** | **46.0** |
|
| 108 |
|
| 109 |
## 3. Usage
|
| 110 |
|