List your company

Theorem Proving on MO-INT (Proof Length)

17Proof LengthClaude 4.6 Opus
12.37643.58874.8106.012Apr 29, 2026

Evaluation results

MethodLinks
Claude 4.6 OpusType=Proprietary LLM2026.0417
Gemini 2.5 ProType=Proprietary LLM2026.0435
DreamProverBackbone=Gemini 2.5 Pro, Type=Lemma Learning System2026.0445.3
Gemini 3.1 ProType=Proprietary LLM2026.0447.9
DreamProverBackbone=Gemini 3.1 Pro, Type=Lemma Learning System2026.0450.6
DreamProverBackbone=GPT-5.3-Codex, Type=Lemma Learning System2026.0455.3
GPT-5.3-CodexType=Proprietary LLM2026.0460.5
DeepSeek-Prover-V2-7BType=Open-source LLM2026.0461.8
Goedel-Prover-V2-8BType=Open-source LLM2026.0476
HilbertBackbone=Gemini 2.5 Pro, Type=Agentic System2026.0493.4
HilbertBackbone=GPT-5.3-Codex, Type=Agentic System2026.04124.1
Goedel-Prover-V2-32BType=Open-source LLM2026.04128
HilbertBackbone=Gemini 3.1 Pro, Type=Agentic System2026.04132.6