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 25.7 #81 of 339
Knowledge 5
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 33.2% #65 of 323+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 63% #117 of 323
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA -8.9 #92 of 323
SimpleQA Verifiedv1 Accuracy on short factual questions with verified answers, answered without web access 19.1% #71 of 84
Mystery Game Puzzles Models choose the best move in puzzle positions from an undisclosed game Epoch 6% #64 of 65
Reasoning 4
ARC-AGI-1public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 89.4% #35 of 64+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 84% #38 of 67
ARC-AGI-2public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 39.3% #34 of 66+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 40.1% #38 of 69
Math 3
AA Analyst Agent Data-analyst tasks: the model explores data with tools and reports findings AA 27.5% #27 of 43
FrontierMath Tier 4 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 17.1% #43 of 55
FrontierMath Tiers 1–3 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 46.3% #49 of 70
Science 4
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 33.3% #74 of 339
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 8.3% #70 of 326
SciCode Writing code for real scientific computing problems (physics, biology, math) AA 49.7% #54 of 142
Terminal-Bench Science0.1 Agents complete expert-authored research workflows across five scientific domains in a terminal AA 0% #36 of 45
Coding 6
Terminal-Bench4.0 Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite 1% #56 of 134+1 versions & slices 2.1 Older, easier version of Terminal-Bench; Still widely reported AA 55.1% #62 of 156
FrontierCode1.1 Agents implement changes in real repositories, graded against maintainer-defined requirements AA 0.6% #37 of 38
WebDev Arena Human pairwise preferences in the stated webdev vote pool Arena 1,409 #65 of 97+2 versions & slices October 23, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,409 #66 of 98October 19, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,409 #65 of 97
Agents 5
AutomationBench-AA Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 4.8% #88 of 135
AA-Briefcase1.1 · Elo Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 912 #67 of 142
EnterpriseOps-Gym-AA Agents complete stateful enterprise workflows in eight domains; final database state determines success AA 42.7% #17 of 43
GDPval-AA (work tasks)2.1 Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work AA 1,133 #57 of 177
τ³-BankingAA Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 18.8% #59 of 134
Multimodal 3
MMMU-Pro (vision) College-level questions that require understanding images: charts, diagrams, photos AA 74% #58 of 128
Arena Vision Human pairwise preferences in the stated vision vote pool Arena 1,237 #52 of 88
Chartography Expert chart reasoning over professional scientific, medical, financial and engineering charts Surge 9.1% #36 of 37
Writing 18
Arena Textcreative writing · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,334 #108 of 197+16 versions & slices creative writing Human pairwise preferences in the stated text vote pool Arena 1,314 #120 of 195overall Human pairwise preferences in the stated text vote pool Arena 1,404 #94 of 196Chinese Human pairwise preferences in the stated text vote pool Arena 1,463 #74 of 198English Human pairwise preferences in the stated text vote pool Arena 1,421 #86 of 205French Human pairwise preferences in the stated text vote pool Arena 1,434 #77 of 164German Human pairwise preferences in the stated text vote pool Arena 1,409 #72 of 167instruction following Human pairwise preferences in the stated text vote pool Arena 1,399 #87 of 205Japanese Human pairwise preferences in the stated text vote pool Arena 1,399 #50 of 152Korean Human pairwise preferences in the stated text vote pool Arena 1,371 #69 of 156non-English Human pairwise preferences in the stated text vote pool Arena 1,401 #84 of 205Polish Human pairwise preferences in the stated text vote pool Arena 1,426 #59 of 142Russian Human pairwise preferences in the stated text vote pool Arena 1,394 #94 of 203Spanish Human pairwise preferences in the stated text vote pool Arena 1,425 #71 of 168overall · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,413 #81 of 188creative writing · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,331 #105 of 187overall · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,413 #83 of 198
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 1,491 #52 of 94
Health & law 6
Harvey LAB-AA1.1 Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit AA 0% #18 of 23
MLCR-AA Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 7.8% #67 of 103
Vals LegalBench Legal reasoning and legal-language understanding on LegalBench Vals 83% #48 of 89
Vals MedCode Medical coding accuracy on patient encounters Vals 37.9% #57 of 78
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 84.1% #33 of 80
Vals Legal Research Bench Grounded legal research; a task passes only if all criteria pass Vals 25.5% #49 of 73
Documents 2
AA-LCR (long context)1.1 Questions that need reading and connecting facts across very long documents (around 100K tokens) AA 75.7% #88 of 317
GDP.pdfAA all-pass Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 11.2% #54 of 135
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 89.5% #41 of 315+1 versions & slices Epoch Models answer graduate-level science multiple-choice questions Epoch 88.5% #41 of 134
OTIS Mock AIME 20242025 Models solve the stated OTIS mock AIME mathematics problems Epoch 90% #48 of 131
No benchmark matches
CheaperDeepSeek V4.1 Flash 1017highOpen+13%-18%$0.30 / $1.20Memory (4-bit): 160 GB → 458 GB+5
BetterGPT-6 Lunamax+3%-69%$0.068Function calling only with none · Chat Completions+3
OpenGLM 5.3 Flashmax+8%-63%$0.28Memory (4-bit): 160 GB → 193 GB+5
CheaperClaude Haiku 5.5max+6%-69%$0.21Remove sampling parameters+5
Qwen3.8 27BxhighOpen+4%+50%$1.15max_tokens range · QwenCloud+4RBeam 1.0 501B A23BxhighOpen-12%-33%$0.24 / $0.98Memory (4-bit): 160 GB → 301 GB+1
MiniMax M2.5unstatedOpen-24%-18%$0.30 / $1.20Context: 256K → 205K+1
Nemotron 3.5 Super VLhighOpen-73%-79%$0.06 / $0.35