List your company

General Knowledge Reasoning on Humanity's Last Exam

50ScoreFugu-Ultra
4.44816.27428.139.926Jun 19, 2026Jun 20, 2026

Evaluation results

MethodLinks
Fugu-Ultra2026.0650
Claude Opus 4.82026.0649.8
Fugu2026.0647.2
Gemini 3.12026.0644.4
GPT-5.52026.0641.4
Confident DecodingModel=Qwen3.5-27B, Decoding Strategy=Confident (Ours)2026.0616.8
Confident DecodingModel=Qwen3.5-122B-A10B, Decoding Strategy=Confident (Ours)2026.0616.5
Last LayerModel=Qwen3.5-27B, Decoding Strategy=Last Layer2026.0616
Last LayerModel=Qwen3.5-122B-A10B, Decoding Strategy=Last Layer2026.0614.7
Confident DecodingModel=gpt-oss-120B, Decoding Strategy=Confident (Ours)2026.0612.6
Confident DecodingModel=Qwen3.5-35B-A3B, Decoding Strategy=Confident (Ours)2026.0611.2
Confident DecodingModel=Gemma-4-31B, Decoding Strategy=Confident (Ours)2026.0611
Last LayerModel=gpt-oss-120B, Decoding Strategy=Last Layer2026.0610.8
Last LayerModel=Qwen3.5-35B-A3B, Decoding Strategy=Last Layer2026.069.2
Last LayerModel=Gemma-4-31B, Decoding Strategy=Last Layer2026.069.2
Confident DecodingModel=gpt-oss-20b, Decoding Strategy=Confident (Ours)2026.066.3
Last LayerModel=gpt-oss-20b, Decoding Strategy=Last Layer2026.066.2