Theorem Proving on PutnamBench (Proof Length Metric)
62.5Average Proof LengthDreamProver
Evaluation Results
| Method | Links | |
|---|---|---|
| DreamProverBackbone=Gemini 2.5 Pro, Type=Lemma Learning System2026.04 | 62.5 | |
| Gemini 3.1 ProType=Proprietary LLM2026.04 | 67.3 | |
| DreamProverBackbone=GPT-5.3-Codex, Type=Lemma Learning System2026.04 | 78.9 | |
| DreamProverBackbone=Gemini 3.1 Pro, Type=Lemma Learning System2026.04 | 83.2 | |
| GPT-5.3-CodexType=Proprietary LLM2026.04 | 85.7 | |
| HilbertBackbone=Gemini 2.5 Pro, Type=Agentic System2026.04 | 126.9 | |
| HilbertBackbone=Gemini 3.1 Pro, Type=Agentic System2026.04 | 129.4 | |
| Goedel-Prover-V2-32BType=Open-source LLM2026.04 | 146.2 | |
| HilbertBackbone=GPT-5.3-Codex, Type=Agentic System2026.04 | 157.3 |