Mathematical Problem Solving on Code2Math
98Accuracy RatioGemini-3-Pro-Thinking
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Gemini-3-Pro-ThinkingEvolver LLM=DeepSeek-Chat, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Open source2026.03 | 98 | 16 | 7 | 9 | |
| Gemini-3-Pro-ThinkingEvolver LLM=DeepSeek-Reasoner, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Open source2026.03 | 98 | 48 | 8 | 40 | |
| Gemini-3-Pro-ThinkingEvolver LLM=Qwen3-235B-A22B-Thinking, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Open source2026.03 | 98 | 20 | 18 | 2 | |
| Gemini-3-Pro-ThinkingEvolver LLM=Gemini-3-Flash-Thinking, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Close source2026.03 | 98 | 56 | 24 | 32 | |
| Gemini-3-Pro-ThinkingEvolver LLM=GPT-5.2-Medium, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Close source2026.03 | 98 | 70 | 52 | 18 | |
| Gemini-3-Pro-ThinkingEvolver LLM=GPT-5.2-High, Solver LLM=Gemini-3-Pro-Thinking, Source Type=Close source2026.03 | 98 | 70 | 61 | 9 | |
| DeepSeek-ReasonerEvolver LLM=DeepSeek-Chat, Solver LLM=DeepSeek-Reasoner, Source Type=Open source2026.03 | 94 | 16 | 9 | 7 | |
| DeepSeek-ReasonerEvolver LLM=DeepSeek-Reasoner, Solver LLM=DeepSeek-Reasoner, Source Type=Open source2026.03 | 94 | 48 | 36 | 12 | |
| DeepSeek-ReasonerEvolver LLM=Qwen3-235B-A22B-Thinking, Solver LLM=DeepSeek-Reasoner, Source Type=Open source2026.03 | 94 | 20 | 19 | 1 | |
| DeepSeek-ReasonerEvolver LLM=Gemini-3-Flash-Thinking, Solver LLM=DeepSeek-Reasoner, Source Type=Close source2026.03 | 94 | 56 | 35 | 21 | |
| DeepSeek-ReasonerEvolver LLM=GPT-5.2-Medium, Solver LLM=DeepSeek-Reasoner, Source Type=Close source2026.03 | 94 | 70 | 55 | 15 | |
| DeepSeek-ReasonerEvolver LLM=GPT-5.2-High, Solver LLM=DeepSeek-Reasoner, Source Type=Close source2026.03 | 94 | 70 | 64 | 6 | |
| DeepSeek-ChatEvolver LLM=DeepSeek-Chat, Solver LLM=DeepSeek-Chat, Source Type=Open source2026.03 | 83 | 16 | 19 | 3 | |
| Seed-2.0-ProEvolver LLM=DeepSeek-Chat, Solver LLM=Seed-2.0-Pro, Source Type=Open source2026.03 | 83 | 16 | 9 | 7 | |
| DeepSeek-ChatEvolver LLM=DeepSeek-Reasoner, Solver LLM=DeepSeek-Chat, Source Type=Open source2026.03 | 83 | 48 | 34 | 14 | |
| Seed-2.0-ProEvolver LLM=DeepSeek-Reasoner, Solver LLM=Seed-2.0-Pro, Source Type=Open source2026.03 | 83 | 48 | 26 | 22 | |
| DeepSeek-ChatEvolver LLM=Qwen3-235B-A22B-Thinking, Solver LLM=DeepSeek-Chat, Source Type=Open source2026.03 | 83 | 20 | 15 | 5 | |
| Seed-2.0-ProEvolver LLM=Qwen3-235B-A22B-Thinking, Solver LLM=Seed-2.0-Pro, Source Type=Open source2026.03 | 83 | 20 | 11 | 9 | |
| DeepSeek-ChatEvolver LLM=Gemini-3-Flash-Thinking, Solver LLM=DeepSeek-Chat, Source Type=Close source2026.03 | 83 | 56 | 33 | 23 | |
| Seed-2.0-ProEvolver LLM=Gemini-3-Flash-Thinking, Solver LLM=Seed-2.0-Pro, Source Type=Close source2026.03 | 83 | 56 | 24 | 32 | |
| DeepSeek-ChatEvolver LLM=GPT-5.2-Medium, Solver LLM=DeepSeek-Chat, Source Type=Close source2026.03 | 83 | 70 | 59 | 11 | |
| Seed-2.0-ProEvolver LLM=GPT-5.2-Medium, Solver LLM=Seed-2.0-Pro, Source Type=Close source2026.03 | 83 | 70 | 51 | 19 | |
| DeepSeek-ChatEvolver LLM=GPT-5.2-High, Solver LLM=DeepSeek-Chat, Source Type=Close source2026.03 | 83 | 70 | 70 | 0 | |
| Seed-2.0-ProEvolver LLM=GPT-5.2-High, Solver LLM=Seed-2.0-Pro, Source Type=Close source2026.03 | 83 | 70 | 61 | 9 | |
| Kimi-K2-ThinkingEvolver LLM=DeepSeek-Chat, Solver LLM=Kimi-K2-Thinking, Source Type=Open source2026.03 | 74 | 16 | 11 | 5 | |
| Kimi-K2-ThinkingEvolver LLM=DeepSeek-Reasoner, Solver LLM=Kimi-K2-Thinking, Source Type=Open source2026.03 | 74 | 48 | 39 | 9 | |
| Kimi-K2-ThinkingEvolver LLM=Qwen3-235B-A22B-Thinking, Solver LLM=Kimi-K2-Thinking, Source Type=Open source2026.03 | 74 | 20 | 14 | 6 | |
| Kimi-K2-ThinkingEvolver LLM=Gemini-3-Flash-Thinking, Solver LLM=Kimi-K2-Thinking, Source Type=Close source2026.03 | 74 | 56 | 28 | 28 | |
| Kimi-K2-ThinkingEvolver LLM=GPT-5.2-Medium, Solver LLM=Kimi-K2-Thinking, Source Type=Close source2026.03 | 74 | 70 | 58 | 12 | |
| Kimi-K2-ThinkingEvolver LLM=GPT-5.2-High, Solver LLM=Kimi-K2-Thinking, Source Type=Close source2026.03 | 74 | 70 | 67 | 3 |