STEM & Reasoning on MathVerse
91.62AccuracyGemini 3-Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini 3-Pro2026.02 | 91.62 | |
| Gemini 2.5-Pro2026.02 | 86.27 | |
| ERNIE 5.02026.02 | 85.13 | |
| GPT-5tier=High2026.02 | 84.19 | |
| Qwen3-VLmode=Thinking2026.02 | 83.96 | |
| CodePercept-32B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 73.56 | |
| Gemini2.5-ProLLM Solver=Qwen3-30A3-Thinking2026.03 | 73.47 | |
| CodePercept-32B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 71.7 | |
| Qwen3-VL-32B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 71.09 | |
| Qwen3-VL-235A22B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 70.08 | |
| Qwen3-VL-32B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 69.9 | |
| GPT5-ThinkingLLM Solver=Qwen3-30A3-Thinking2026.03 | 69.56 | |
| CodePercept-8B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 67.95 | |
| CodePercept-4B-S1LLM Solver=Qwen3-235A22-Thinking2026.03 | 66.73 | |
| CodePercept-8B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 66.52 | |
| Qwen3-VL-30A3B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 66.44 | |
| Qwen3-VL-4B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 66.39 | |
| CodePercept-4B-S1LLM Solver=Qwen3-30A3-Thinking2026.03 | 65.59 | |
| Qwen3-VL-4B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 64.59 | |
| Qwen3-VL-8B-InstructLLM Solver=Qwen3-235A22-Thinking2026.03 | 63.88 | |
| Qwen3-VL-8B-InstructLLM Solver=Qwen3-30A3-Thinking2026.03 | 63.75 | |
| MiniCPM-V-4.5LLM Solver=Qwen3-30A3-Thinking2026.03 | 57.84 | |
| Claude-Opus 4.1-ThinkingLLM Solver=Qwen3-30A3-Thinking2026.03 | 56.19 | |
| Qwen2.5-VL-72BLLM Solver=Qwen3-30A3-Thinking2026.03 | 55.4 | |
| GLM-4.1V-9BLLM Solver=Qwen3-30A3-Thinking2026.03 | 54.47 | |
| InternVL3.5-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 53.4 | |
| Intern-S1-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 51.9 | |
| KeyeVL1.5-8BLLM Solver=Qwen3-30A3-Thinking2026.03 | 49.95 |