Theorem Proving on MO-INT (Proof Length)
17Proof LengthClaude 4.6 Opus
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude 4.6 OpusType=Proprietary LLM2026.04 | 17 | |
| Gemini 2.5 ProType=Proprietary LLM2026.04 | 35 | |
| DreamProverBackbone=Gemini 2.5 Pro, Type=Lemma Learning System2026.04 | 45.3 | |
| Gemini 3.1 ProType=Proprietary LLM2026.04 | 47.9 | |
| DreamProverBackbone=Gemini 3.1 Pro, Type=Lemma Learning System2026.04 | 50.6 | |
| DreamProverBackbone=GPT-5.3-Codex, Type=Lemma Learning System2026.04 | 55.3 | |
| GPT-5.3-CodexType=Proprietary LLM2026.04 | 60.5 | |
| DeepSeek-Prover-V2-7BType=Open-source LLM2026.04 | 61.8 | |
| Goedel-Prover-V2-8BType=Open-source LLM2026.04 | 76 | |
| HilbertBackbone=Gemini 2.5 Pro, Type=Agentic System2026.04 | 93.4 | |
| HilbertBackbone=GPT-5.3-Codex, Type=Agentic System2026.04 | 124.1 | |
| Goedel-Prover-V2-32BType=Open-source LLM2026.04 | 128 | |
| HilbertBackbone=Gemini 3.1 Pro, Type=Agentic System2026.04 | 132.6 |