List your company

Mathematics Reasoning on IMO-Ans

80.6AccuracyVibeThinker-3B
26.62440.63754.6568.663Jun 15, 2026

Evaluation results

MethodLinks
VibeThinker-3BParams=3B, CLR reasoning enhancement=true2026.0680.6
VibeThinker-3BParams=3B2026.0676.4
Qwen3-235B-A22B-ThinkingParams=235B2026.0670.5
Nemotron-3-NanoParams=30B2026.0670.4
Ministral-3-Reasoning-2512Params=14B2026.0663.4
GPT-OSS-20B (high)Params=20B2026.0661.9
OpenReasoning-NemotronParams=7B2026.0660.6
Gemma-4-itParams=12B2026.0654.9
Mimo7B-RL-0530Params=7B2026.0653.9
Qwen3-4B-Thinking-2507Params=4B2026.0651.6
Olmo-3-ThinkParams=7B2026.0649.4
Qwen3.5-4BParams=4B2026.0648.7
Phi4-Reasoning-PlusParams=14B2026.0646.2
Hunyuan-4B-InstructParams=4B2026.0639.6
SmolLM3Params=3B2026.0628.7