Omni-modal Reasoning on MathVerse mini (test)
59.3ScoreGemini-2.0-Flash
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-2.0-Flash2026.02 | 59.3 | |
| GPT-4o2026.02 | 50.8 | |
| ThinkOmniBase Model=Omni-R1-7B, LRM Guide=Qwen32026.02 | 45.2 | |
| ThinkOmniBase Model=Qwen2.5-Omni-7B, LRM Guide=Qwen32026.02 | 45.1 | |
| ThinkOmniBase Model=Omni-R1-7B, LRM Guide=DeepSeek2026.02 | 43.1 | |
| ThinkOmniBase Model=Qwen2.5-Omni-7B, LRM Guide=DeepSeek2026.02 | 42 | |
| Qwen2.5-Omni-7B2026.02 | 40.2 | |
| Omni-R1Notes=Evaluated using authors' scripts2026.02 | 39.8 | |
| Omni-R12026.02 | 39.8 | |
| ThinkOmniBase Model=Qwen2.5-Omni-3B, LRM Guide=Qwen32026.02 | 38.8 | |
| HumanOmniV2Notes=Evaluated using authors' scripts2026.02 | 37.3 | |
| ThinkOmniBase Model=Qwen2.5-Omni-3B, LRM Guide=DeepSeek2026.02 | 33.5 | |
| Qwen2.5-Omni-3B2026.02 | 32 |