STEM Reasoning on MathVision
69.96AccuracyCodePercept-32B-S1
Evaluation Results
| Method | Links | |
|---|---|---|
| CodePercept-32B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 69.96 | |
| Gemini2.5-ProLLM Solver=Qwen3-30A3-Thinking2026.03 | 66.8 | |
| CodePercept-8B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 66.45 | |
| CodePercept-4B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 64.71 | |
| Qwen3-VL-32B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 62.66 | |
| CodePercept-32B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 62.27 | |
| Qwen3-VL-235A22B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 60.43 | |
| GPT5-ThinkingLLM Solver=Qwen3-30A3-Thinking2026.03 | 60.03 | |
| Qwen3-VL-4B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 59.8 | |
| Qwen3-VL-8B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 59.67 | |
| Claude-Opus 4.1-ThinkingLLM Solver=Qwen3-30A3-Thinking2026.03 | 59.61 | |
| CodePercept-8B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 59.31 | |
| Qwen3-VL-32B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 58.55 | |
| CodePercept-4B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 57.63 | |
| Qwen3-VL-8B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 54.37 | |
| Qwen3-VL-4B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 54.21 | |
| Qwen2.5-VL-72BLLM Solver=Qwen3-30A3-Thinking2026.03 | 54.14 | |
| KeyeVL1.5-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 54.11 | |
| GLM-4.1V-9BLLM Solver=Qwen3-30A3-Thinking2026.03 | 53.75 | |
| Qwen3-VL-30A3B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 53.59 | |
| InternVL3.5-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 53.32 | |
| MiniCPM-V-4.5LLM Solver=Qwen3-30A3-Thinking2026.03 | 53.15 | |
| Intern-S1-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 51.67 |