benchgap
Calibration

AA-Omniscience Accuracy → SuperGPQA

SuperGPQA is estimated from AA-Omniscience Accuracy with a Hill curve fitted on 12 models measured on both: y = 0.6435 + (1.2000 − 0.6435)·x^6.00 / (0.45938^6.00 + x^6.00), R² = 0.59, cross-validated error 9.3 pp. It is used for 84 estimates.

Estimated modelAA-Omniscience AccuracySuperGPQASource
Claude Fable 565.4%100.0%estimated ± 9.3 pp, low confidence
Claude Opus 5.565.5%100.0%estimated ± 9.3 pp, low confidence
Claude Sonnet 5.551.5%100.0%estimated ± 9.3 pp, low confidence
Gemini 3.7 Flash55.3%100.0%estimated ± 9.3 pp, low confidence
Gemini 3 Pro55.8%100.0%estimated ± 9.3 pp, low confidence
Gemini 4 Carbon58.3%100.0%estimated ± 9.3 pp, low confidence
GPT-6.1 Sol61.8%100.0%estimated ± 9.3 pp, low confidence
GPT-6 Sol54.5%100.0%estimated ± 9.3 pp, low confidence
Grok 4.551.6%100.0%estimated ± 9.3 pp, low confidence
Gemini 3.8 Flash50.5%99.9%estimated ± 9.3 pp, low confidence
Gemini 4 Hydrogen50.5%99.9%estimated ± 9.3 pp, low confidence
Gemini 3.6 Flash50.0%99.1%estimated ± 9.3 pp, low confidence
Gemini 4 Argon48.3%96.3%estimated ± 9.3 pp, low confidence
Grok 4.648.2%96.2%estimated ± 9.3 pp, low confidence
GPT-6.1 Terra47.6%95.1%estimated ± 9.3 pp, low confidence
Claude Opus 4.6 (Adaptive)47.0%94.1%estimated ± 9.3 pp, low confidence
Claude Opus 4.5 Thinking46.6%93.4%estimated ± 9.3 pp, low confidence
Gemini 3 Flash45.8%91.9%estimated ± 9.3 pp, medium confidence
Muse Spark 1.245.4%91.2%estimated ± 9.3 pp, medium confidence
MiniMax M3.1 Flash45.1%90.6%estimated ± 9.3 pp, medium confidence
MiniMax M344.9%90.3%estimated ± 9.3 pp, medium confidence
Muse Spark 1.343.6%87.8%estimated ± 9.3 pp, medium confidence
Grok 4.742.7%86.2%estimated ± 9.3 pp, medium confidence
Muse Spark Open42.0%84.9%estimated ± 9.3 pp, medium confidence
GPT-5.2-Codex41.1%83.2%estimated ± 9.3 pp, medium confidence
Muse Wonder Full41.0%83.0%estimated ± 9.3 pp, medium confidence
Grok 440.5%82.1%estimated ± 9.3 pp, medium confidence
GPT-540.3%81.8%estimated ± 9.3 pp, medium confidence
GPT-5.1-Codex39.9%81.1%estimated ± 9.3 pp, medium confidence
GPT-5.1-Codex-Max39.9%81.1%estimated ± 9.3 pp, medium confidence
GPT-6 Luna39.6%80.5%estimated ± 9.3 pp, medium confidence
Muse Wonder38.0%77.8%estimated ± 9.3 pp, medium confidence
Ling 3.0 Tiny37.9%77.7%estimated ± 9.3 pp, medium confidence
GPT-5.137.7%77.4%estimated ± 9.3 pp, medium confidence
MiMo-V2.6-Pro34.8%73.2%estimated ± 9.3 pp, medium confidence
Claude Haiku 5.534.7%73.1%estimated ± 9.3 pp, medium confidence
GLM-5.332.3%70.3%estimated ± 9.3 pp, medium confidence
Beam 1.0 501B-A23B30.7%68.9%estimated ± 9.3 pp, medium confidence
Qwen3.8 Max Preview29.5%68.0%estimated ± 9.3 pp, medium confidence
GLM-5V-Turbo29.3%67.9%estimated ± 9.3 pp, medium confidence
GLM-5-Turbo28.4%67.3%estimated ± 9.3 pp, medium confidence
Apodex 1.1 Mini28.2%67.2%estimated ± 9.3 pp, medium confidence
Muse Glimmer 30B28.0%67.1%estimated ± 9.3 pp, medium confidence
MiMo-V2.6-Flash27.2%66.6%estimated ± 9.3 pp, medium confidence
MiMo-V2-Pro26.6%66.4%estimated ± 9.3 pp, medium confidence
Gemini 3.5 Flash-Lite26.5%66.3%estimated ± 9.3 pp, medium confidence
Gemini 2.5 Flash26.1%66.2%estimated ± 9.3 pp, medium confidence
Mistral Large 426.0%66.1%estimated ± 9.3 pp, medium confidence
Step 3.7 Flash25.8%66.0%estimated ± 9.3 pp, medium confidence
Grok 4.1 Fast (Reasoning)25.1%65.8%estimated ± 9.3 pp, medium confidence
Mistral Large 325.0%65.8%estimated ± 9.3 pp, medium confidence
Qwen3 Max24.4%65.6%estimated ± 9.3 pp, medium confidence
DeepSeek V3 032424.3%65.5%estimated ± 9.3 pp, medium confidence
DeepSeek V3.224.0%65.5%estimated ± 9.3 pp, medium confidence
Grok Code Fast 123.5%65.3%estimated ± 9.3 pp, medium confidence
Mistral Medium 3.5 128B23.0%65.2%estimated ± 9.3 pp, medium confidence
Grok 4 Fast (Reasoning)22.8%65.2%estimated ± 9.3 pp, medium confidence
Claude 4 Sonnet22.7%65.1%estimated ± 9.3 pp, medium confidence
Mercury 2.522.0%65.0%estimated ± 9.3 pp, medium confidence
Nemotron 3.5 Bolt 74B-A9B BF1621.8%65.0%estimated ± 9.3 pp, medium confidence
GLM-4.621.4%64.9%estimated ± 9.3 pp, medium confidence
Mistral Small 421.4%64.9%estimated ± 9.3 pp, medium confidence
Muse Ember21.1%64.9%estimated ± 9.3 pp, medium confidence
GPT-4o19.9%64.7%estimated ± 9.3 pp, medium confidence
Mistral Large 219.9%64.7%estimated ± 9.3 pp, medium confidence
MiMo-V2-Omni19.3%64.7%estimated ± 9.3 pp, low confidence
Solar Pro 318.5%64.6%estimated ± 9.3 pp, low confidence
North Mini Code18.4%64.6%estimated ± 9.3 pp, low confidence
Solar Pro 418.4%64.6%estimated ± 9.3 pp, low confidence
Mistral Medium 318.3%64.6%estimated ± 9.3 pp, low confidence
Claude 3 Haiku17.6%64.5%estimated ± 9.3 pp, low confidence
Grok 4.1 Fast17.2%64.5%estimated ± 9.3 pp, low confidence
Nova Pro16.9%64.5%estimated ± 9.3 pp, low confidence
K-Exaone16.4%64.5%estimated ± 9.3 pp, low confidence
GLM-4.5-Air16.3%64.5%estimated ± 9.3 pp, low confidence
Solar Pro 216.1%64.4%estimated ± 9.3 pp, low confidence
Mistral Small 3.115.1%64.4%estimated ± 9.3 pp, low confidence
Devstral Small14.7%64.4%estimated ± 9.3 pp, low confidence
GPT-4.1 nano13.7%64.4%estimated ± 9.3 pp, low confidence
Claude 3.5 Haiku13.2%64.4%estimated ± 9.3 pp, low confidence
Gemma 3 27B13.0%64.4%estimated ± 9.3 pp, low confidence
Command A+9.8%64.4%estimated ± 9.3 pp, low confidence
LFM2.5-8B-A1B9.4%64.4%estimated ± 9.3 pp, low confidence
Granite 4.2 3B9.2%64.3%estimated ± 9.3 pp, low confidence

Fitted on

ModelAA-Omniscience AccuracySuperGPQA
Claude Opus 4.540.9%70.6%
Claude Opus 4.645.8%95.0%
Claude Sonnet 4.638.6%95.0%
GLM-526.3%66.8%
Kimi K227.4%57.2%
Kimi K2.535.2%69.2%
Qwen3.5-27B20.7%65.6%
Qwen3.5-35B-A3B20.1%63.4%
Qwen3.6-27B19.6%66.0%
Qwen 3.6 Max (preview)37.9%73.9%
Qwen3.6 Plus26.4%71.6%
Qwen3.7 Max31.1%73.6%