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 32.9 #51 of 339
Knowledge 5
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 42.8% #41 of 323+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 84.4% #199 of 323
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA -5.5 #86 of 323
SimpleQA Verifiedv1 Accuracy on short factual questions with verified answers, answered without web access 41.4% #39 of 84
Mystery Game Puzzles Models choose the best move in puzzle positions from an undisclosed game Epoch 7% #61 of 65
Reasoning 6
ARC-AGI-1public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 79.8% #38 of 64+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 70.3% #42 of 67
ARC-AGI-2public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 31.9% #37 of 66+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 31.4% #40 of 69
ARC-AGI-3semi-private Measures interactive game solving through relative human action efficiency ARC 0.2% #18 of 19
Chess Puzzles Models choose the best next move from a novel chess position Epoch 36% #37 of 69
Math 3
AA Analyst Agent Data-analyst tasks: the model explores data with tools and reports findings AA 43.8% #17 of 43
FrontierMath Tier 4 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 56.1% #17 of 55
FrontierMath Tiers 1–3 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 78.9% #18 of 70
Science 5
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 32.9% #74 of 339
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 15.4% #44 of 326
Vals Science Index Scientific problem solving across the index’s component evaluations Vals 28.4% #16 of 19
SciCode Writing code for real scientific computing problems (physics, biology, math) AA 50.3% #51 of 142
Terminal-Bench Science0.1 Agents complete expert-authored research workflows across five scientific domains in a terminal AA 2.9% #30 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 4.5% #46 of 134
DeepSWE1.1 Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition 67% #17 of 44
FrontierCode1.1 Agents implement changes in real repositories, graded against maintainer-defined requirements AA 32% #27 of 38
WebDev Arena Human pairwise preferences in the stated webdev vote pool Arena 1,578 #30 of 97+2 versions & slices October 23, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,579 #31 of 98October 19, 2026 Human pairwise preferences in the stated webdev vote pool Arena 1,578 #30 of 97
Agents 7
AutomationBench-AA Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 47.8% #40 of 135
AA-Briefcase1.1 · Elo Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 1,186 #45 of 142
GDPval-AA (work tasks)2.1 Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work AA 1,352 #44 of 177
Furniture Assembly Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step Epoch 44.2% #13 of 31
SpireClimb Agents play Slay the Spire under the named game and difficulty setting Epoch 0% #7 of 7
SpireClimb StS1 A20 Heart Agents play Slay the Spire under the named game and difficulty setting Epoch 0% #4 of 7
SpireClimb StS2 A10 Agents play Slay the Spire under the named game and difficulty setting Epoch 0% #7 of 7
Multimodal 3
MMMU-Pro (vision) College-level questions that require understanding images: charts, diagrams, photos AA 77.5% #36 of 128
Arena Vision Human pairwise preferences in the stated vision vote pool Arena 1,216 #59 of 88
Chartography Expert chart reasoning over professional scientific, medical, financial and engineering charts Surge 29.1% #15 of 37
Writing 17
Arena Textcreative writing · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,363 #91 of 197+16 versions & slices creative writing Human pairwise preferences in the stated text vote pool Arena 1,410 #56 of 195overall Human pairwise preferences in the stated text vote pool Arena 1,443 #59 of 196Chinese Human pairwise preferences in the stated text vote pool Arena 1,426 #100 of 198English Human pairwise preferences in the stated text vote pool Arena 1,394 #107 of 205French Human pairwise preferences in the stated text vote pool Arena 1,406 #97 of 164German Human pairwise preferences in the stated text vote pool Arena 1,376 #87 of 167instruction following Human pairwise preferences in the stated text vote pool Arena 1,407 #78 of 205Japanese Human pairwise preferences in the stated text vote pool Arena 1,369 #73 of 152Korean Human pairwise preferences in the stated text vote pool Arena 1,364 #75 of 156non-English Human pairwise preferences in the stated text vote pool Arena 1,381 #102 of 205Polish Human pairwise preferences in the stated text vote pool Arena 1,383 #86 of 142Russian Human pairwise preferences in the stated text vote pool Arena 1,393 #95 of 203Spanish Human pairwise preferences in the stated text vote pool Arena 1,391 #97 of 168overall · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,392 #96 of 188creative writing · October 15, 2026 Human pairwise preferences in the stated text vote pool Arena 1,363 #88 of 187overall · October 22, 2026 Human pairwise preferences in the stated text vote pool Arena 1,390 #102 of 198
Health & law 5
Harvey LAB-AA1.1 Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit AA 3.3% #12 of 23
MLCR-AA Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 19.4% #25 of 103
Vals MedCode Medical coding accuracy on patient encounters Vals 44.7% #33 of 78
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 83.7% #37 of 80
Vals Legal Research Bench Grounded legal research; a task passes only if all criteria pass Vals 30.3% #42 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 79.3% #54 of 317
GDP.pdfAA all-pass Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 18.2% #30 of 135
Legacy · saturated or no longer reported 2
GPQA DiamondEpoch Models answer graduate-level science multiple-choice questions Epoch 90.5% #32 of 134
OTIS Mock AIME 20242025 Models solve the stated OTIS mock AIME mathematics problems Epoch 98.9% #16 of 131
No benchmark matches
OpenKimi K3max+21%+6,790%$2.00Context: 1.05M → 1M+4
CheaperClaude Haiku 5.5low-38%-16%$0.024Context: 1.05M → 1M+6
CheaperLing 3.0 Flash VLthinkingOpen-54%-55%$0.06 / $0.18Context: 1.05M → 262K
MiMo v2.6 ProthinkingOpen+5%+359%$0.13Context: 1.05M → 1M+4
Muse Spark OpenxhighOpensame+2,808%$0.84Context: 1.05M → 1M+2
DeepSeek V4.1 Flash 1017maxOpen-5%+803%$0.26Context: 1.05M → 1M+5
Qwen3.8 27BxhighOpen-10%+3,849%$1.15Context: 1.05M → 262K+5