List your company

Reasoning on HMMT

85Pass@1Vanilla
8.580828.420448.2668.0996Dec 8, 2025Dec 18, 2025Dec 28, 2025Jan 8, 2026Jan 18, 2026Jan 28, 2026Feb 8, 2026

Evaluation results

MethodLinks
VanillaBackbone=Qwen3-235B2026.028529,100
Free()LMBackbone=Qwen3-235B2026.0284.5821,700
GeminiBackbone=Qwen3-235B2026.0284.1725,600
No TrainBackbone=Qwen3-235B2026.028024,400
Free()LMBackbone=Qwen3-30B2026.0270.4219,100
GeminiBackbone=Qwen3-30B2026.0265.4219,000
No TrainBackbone=Qwen3-30B2026.0263.3320,400
VanillaBackbone=Qwen3-30B2026.0260.8321,000
Free()LMBackbone=Qwen3-8B2026.0249.5816,400
CompassMax-V3-Thinking2025.1246.7—
H2OBackbone=Qwen3-8B2026.0246.6721,500
GeminiBackbone=Qwen3-8B2026.0244.1718,400
No TrainBackbone=Qwen3-8B2026.0242.519,900
DeepSeek-R12025.1241.67—
VanillaBackbone=Qwen3-8B2026.0238.7519,400
TCBackbone=Qwen3-8B2026.0226.6720,500
CompassMax-V32025.1211.52—