Benchmark By Score Rank
Overall 1
AA Intelligence Index4.3.2 Artificial Analysis combines tests of knowledge, reasoning, coding and tool use into one score AA 38.8 #35 of 334
Knowledge 17
Arena Text Human pairwise preferences in the stated text vote pool Arena 1,466 #34 of 196+12 versions & slices English Human pairwise preferences in the stated text vote pool Arena 1,453 #44 of 188instruction following Human pairwise preferences in the stated text vote pool Arena 1,446 #38 of 188non-English Human pairwise preferences in the stated text vote pool Arena 1,440 #36 of 188overall · full cohort Human pairwise preferences in the stated text vote pool Arena 1,448 #38 of 188Russian Human pairwise preferences in the stated text vote pool Arena 1,449 #37 of 186Chinese Human pairwise preferences in the stated text vote pool Arena 1,495 #44 of 181Spanish Human pairwise preferences in the stated text vote pool Arena 1,451 #41 of 155German Human pairwise preferences in the stated text vote pool Arena 1,448 #38 of 154French Human pairwise preferences in the stated text vote pool Arena 1,457 #46 of 149Korean Human pairwise preferences in the stated text vote pool Arena 1,406 #38 of 144Japanese Human pairwise preferences in the stated text vote pool Arena 1,431 #27 of 138Polish Human pairwise preferences in the stated text vote pool Arena 1,433 #45 of 133
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 25.3 #20 of 318
SimpleQA Verifiedv1 Accuracy on short factual questions with verified answers, answered without web access 53.8% #20 of 83
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 51.6% #17 of 318+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 54.1% #97 of 318
Reasoning 6
Chess Puzzles Models choose the best next move from a novel chess position Epoch 35% #38 of 68
ARC-AGI-2semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 33.1% #37 of 66+1 versions & slices public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 36.3% #33 of 63
ARC-AGI-1semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 79.2% #36 of 64+1 versions & slices public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 86% #34 of 61
ARC-AGI-3semi-private Measures interactive game solving through relative human action efficiency ARC 0.3% #15 of 18
Math 4
FrontierMath Tiers 1–3 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 57.2% #38 of 69
FrontierMath Tier 4 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 24.4% #37 of 54
MathArenaArXivMath · June 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 66.7% #16 of 24+1 versions & slices BrokenArXiv · June 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 29.3% #16 of 24
Science 2
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 42.7% #33 of 334
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 15.4% #43 of 321
Health & law 5
Vals LegalBench Legal reasoning and legal-language understanding on LegalBench Vals 86% #15 of 89
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 86.9% #18 of 80
Vals MedCode Medical coding accuracy on patient encounters Vals 43.3% #38 of 78
Vals Legal Research Bench Grounded legal research; a task passes only if all criteria pass Vals 38% #32 of 69
MLCR-AA Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 15% #34 of 64
Documents 2
AA-LCR (long context)1.1 Questions that need reading and connecting facts across very long documents (around 100K tokens) AA 79.3% #53 of 273
GDP.pdfAA all-pass Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 18.8% #27 of 133
Legacy · saturated or no longer reported 3
GPQA Diamond Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 93.1% #12 of 312+1 versions & slices Epoch Models answer graduate-level science multiple-choice questions Epoch 93.4% #13 of 133
OTIS Mock AIME 20242025 Models solve the stated OTIS mock AIME mathematics problems Epoch 97.8% #26 of 130
No benchmark matches
CheaperGemini 3.8 Flashhigh×39.9-50%*$1.24Sampling parameters are ignored+1
BetterGPT-6.1 Sollow×15.1+33%$0.13Remove unsupported parameters+5
CheaperClaude Haiku 5.5maxsame-93%$0.21Remove sampling parameters+5