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 22.2 #105 of 339
Knowledge 5
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 29.5% #83 of 323+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 34.4% #48 of 323
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 5.2 #52 of 323
SimpleQA Verifiedv1 Accuracy on short factual questions with verified answers, answered without web access 33.1% #55 of 84
Mystery Game Puzzles Models choose the best move in puzzle positions from an undisclosed game Epoch 19% #39 of 65
Reasoning 5
ARC-AGI-1public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 47% #54 of 64+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 32.3% #56 of 67
ARC-AGI-2public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 1.9% #54 of 66+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 5.3% #51 of 69
Chess Puzzles Models choose the best next move from a novel chess position Epoch 21% #50 of 69
Math 2
FrontierMath Tier 4 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 0% #54 of 55
FrontierMath Tiers 1–3 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 26% #60 of 70
Science 4
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 18.8% #130 of 339
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 0% #182 of 326
SciCode Writing code for real scientific computing problems (physics, biology, math) AA 41.3% #92 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 5
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 53.6% #65 of 156
WebDev Arena Human pairwise preferences in the stated webdev vote pool Arena 1,435 #58 of 97+2 versions & slices October 23, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,434 #61 of 98October 19, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,435 #60 of 97
Agents 5
AutomationBench-AA Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 25% #55 of 135
AA-Briefcase1.1 · Elo Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 651 #90 of 142
EnterpriseOps-Gym-AA Agents complete stateful enterprise workflows in eight domains; final database state determines success AA 42.3% #18 of 43
GDPval-AA (work tasks)2.1 Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work AA 995 #73 of 177
τ³-BankingAA Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 17.5% #61 of 134
Multimodal 3
MMMU-Pro (vision) College-level questions that require understanding images: charts, diagrams, photos AA 79% #27 of 128
Arena Vision Human pairwise preferences in the stated vision vote pool Arena 1,269 #32 of 88
Chartography Expert chart reasoning over professional scientific, medical, financial and engineering charts Surge 10.2% #35 of 37
Writing 18
Arena Textcreative writing · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,421 #46 of 197+16 versions & slices creative writing Human pairwise preferences in the stated text vote pool Arena 1,436 #40 of 195overall Human pairwise preferences in the stated text vote pool Arena 1,455 #46 of 196Chinese Human pairwise preferences in the stated text vote pool Arena 1,467 #68 of 198English Human pairwise preferences in the stated text vote pool Arena 1,438 #69 of 205French Human pairwise preferences in the stated text vote pool Arena 1,450 #60 of 164German Human pairwise preferences in the stated text vote pool Arena 1,454 #34 of 167instruction following Human pairwise preferences in the stated text vote pool Arena 1,419 #67 of 205Japanese Human pairwise preferences in the stated text vote pool Arena 1,426 #32 of 152Korean Human pairwise preferences in the stated text vote pool Arena 1,407 #38 of 156non-English Human pairwise preferences in the stated text vote pool Arena 1,427 #55 of 205Polish Human pairwise preferences in the stated text vote pool Arena 1,437 #44 of 142Russian Human pairwise preferences in the stated text vote pool Arena 1,444 #43 of 203Spanish Human pairwise preferences in the stated text vote pool Arena 1,442 #56 of 168overall · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,435 #59 of 188creative writing · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,420 #45 of 187overall · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,435 #61 of 198
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 1,559 #49 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 10.6% #54 of 103
Vals LegalBench Legal reasoning and legal-language understanding on LegalBench Vals 84.1% #35 of 89
Vals MedCode Medical coding accuracy on patient encounters Vals 43.5% #36 of 78
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 70.9% #68 of 80
Vals Legal Research Bench Grounded legal research; a task passes only if all criteria pass Vals 13.9% #62 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 76% #86 of 317
GDP.pdfAA all-pass Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 13.6% #43 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 83.8% #91 of 315+1 versions & slices Epoch Models answer graduate-level science multiple-choice questions Epoch 83.3% #61 of 134
OTIS Mock AIME 20242025 Models solve the stated OTIS mock AIME mathematics problems Epoch 71.1% #75 of 131
No benchmark matches
CheaperDeepSeek V4.1 Flash 1017highOpen×2.9-38%$0.30 / $1.20No PDF input+5
BetterGPT-6 Lunalow+16%-76%$0.005Function calling only with none · Chat Completions+3
CheaperGLM 5.3 FlashmaxOpen×2.8-72%$0.28Defaults to max · Self-hosted+4
CheaperClaude Haiku 5.5max×2.8-76%$0.21Remove sampling parameters+5
Kimi K3maxOpen×2.9+606%$2.00Sampling parameters are fixed+3
Muse Spark OpenxhighOpen×2.8+135%$0.84Reasoning off leaves output unchanged · Self-hosted / llama.cpp+1