Benchmark By Score Rank
Overall 1
AA Intelligence Index4.3.2 AA Intelligence Index 4.3.2: Artificial Analysis combines tests of knowledge, reasoning, coding and tool use into one score AA 53.6 #3 of 337
Knowledge 5
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 64.6% #3 of 321+1 versions & slices Hallucination Rate AA Omniscience · Hallucination Rate: Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 67.6% #126 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 40.6 #6 of 321
SimpleQA Verifiedv1 SimpleQA Verified v1: Accuracy on short factual questions with verified answers, answered without web access 72.2% #6 of 83
Mystery Game Puzzles Mystery Game Puzzles: Models choose the best move in puzzle positions from an undisclosed game Epoch 71% #3 of 64
Reasoning 5
ARC-AGI-1public evaluation ARC-AGI-1 · public-eval: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 98.3% #7 of 61+1 versions & slices semi-private ARC-AGI-1 · semi-private: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 98.5% #1 of 64
ARC-AGI-2public evaluation ARC-AGI-2 · public-eval: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 93.5% #8 of 63+1 versions & slices semi-private ARC-AGI-2 · semi-private: Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 93.3% #3 of 66
Chess Puzzles Chess Puzzles: Models choose the best next move from a novel chess position Epoch 67% #3 of 68
Math 13
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings AA 56.3% #3 of 43
FrontierMath Erdos FrontierMath Erdos: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 2.9% #3 of 13
FrontierMath Tier 4 (v2) FrontierMath Tier 4 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 95% #3 of 54
FrontierMath Tiers 1–3 (v2) FrontierMath Tiers 1–3 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 91.2% #3 of 69
MathArenaArXivMath · August 2026 MathArena · arxiv · 08/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 86% #5 of 12+8 versions & slices ArXivMath · June 2026 MathArena · arxiv · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 83.3% #8 of 24ArXivMath · May 2026 MathArena · arxiv · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 81.7% #6 of 19ArXivMath · September 2026 MathArena · arxiv · 09/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 82.1% #3 of 7BrokenArXiv · August 2026 MathArena · arxiv_false · 08/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 77.4% #5 of 12BrokenArXiv · June 2026 MathArena · arxiv_false · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 24BrokenArXiv · May 2026 MathArena · arxiv_false · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 95% #1 of 18BrokenArXiv · September 2026 MathArena · arxiv_false · 09/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 76.7% #2 of 7ArXivLean · September 2026 MathArena · arxivlean · 09/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 47.2% #1 of 5
Science 5
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 55.6% #3 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 30.9% #7 of 324
Vals Science Index Vals Science Index: Scientific problem solving across the index’s component evaluations Vals 58.7% #4 of 18
SciCode SciCode: Writing code for real scientific computing problems (physics, biology, math) AA 60.4% #7 of 140
Terminal-Bench Science0.1 Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal AA 49% #5 of 33
Coding 7 Leads on 3
Terminal-Bench4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite 56.6% #5 of 132
DeepSWE1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition 72.3% #5 of 44
FrontierCode1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements AA 54% #1 of 28
ITBench SRE ITBench SRE: IT incident tasks: diagnose and fix broken systems like a site-reliability engineer AA 38.2% #20 of 37
WebDev Arena WebDev Arena: Human pairwise preferences in the stated webdev vote pool Arena 1,807 #1 of 97+1 versions & slices full cohort WebDev Arena · snapshot 2026-10-19 · native board settings: Human pairwise preferences in the stated webdev vote pool Arena 1,807 #1 of 97
MirrorCode MirrorCode: Agents reimplement software projects by observing their behavior Epoch 77.4% #1 of 9
Agents 8 Leads on 1
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 63.2% #12 of 133
AA-Briefcase1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 1,686 #3 of 140
GDPval-AA (work tasks)2.1 GDPval-AA (work tasks) 2.1: Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work AA 60.2 #8 of 175
EBR Bench EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs Epoch 71.4% #2 of 24
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step Epoch 83.3% #1 of 31
SpireClimb SpireClimb: Agents play Slay the Spire under the named game and difficulty setting Epoch 3.3% #4 of 7
SpireClimb StS1 A20 Heart SpireClimb StS1 A20 Heart: Agents play Slay the Spire under the named game and difficulty setting Epoch 0% #4 of 7
SpireClimb StS2 A10 SpireClimb StS2 A10: Agents play Slay the Spire under the named game and difficulty setting Epoch 6.7% #4 of 7
Multimodal 3
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 85.8% #3 of 127
Arena Vision Arena Vision · overall: Human pairwise preferences in the stated vision vote pool Arena 1,321 #2 of 88
Chartography Chartography: Expert chart reasoning over professional scientific, medical, financial and engineering charts Surge 66.3% #3 of 37
Writing 14 Leads on 1
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,519 #1 of 195+10 versions & slices overall Arena Text · overall: Human pairwise preferences in the stated text vote pool Arena 1,508 #3 of 196Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool Arena 1,584 #3 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,517 #2 of 188French Arena Text · french: Human pairwise preferences in the stated text vote pool Arena 1,505 #5 of 149instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,536 #1 of 188non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,507 #3 of 188Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,519 #3 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,519 #3 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,516 #2 of 188creative writing · full cohort Arena Text · creative writing · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,534 #1 of 187
EQ-Bench Longform Creative Writing1.11 Longform Creative Writing 1.11: Models plan, revise and write a story across eight turns, graded with a writing rubric EQ-Bench 84.6% #3 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 2,050 #7 of 91
Lech Mazur Short-Story Lech Mazur Short-Story · comparison score: Judges compare stories written to ten required elements; the comparison score can be negative Lech 3.7 #3 of 44
Health & law 5 Leads on 1
Harvey LAB-AA1.1 Harvey LAB-AA 1.1: Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit AA 4.2% #9 of 23
MLCR-AA MLCR-AA: Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 66.7% #3 of 64
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 49.8% #15 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 91.4% #1 of 80
Vals Legal Research Bench Vals Legal Research Bench: Grounded legal research; a task passes only if all criteria pass Vals 50.5% #5 of 69
Documents 2
AA-LCR (long context)1.1 AA-LCR (long context) 1.1: Questions that need reading and connecting facts across very long documents (around 100K tokens) AA 82.7% #21 of 275
GDP.pdfAA all-pass GDP.pdf · AA all-pass: Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 28.8% #5 of 133
Legacy · saturated or no longer reported 2
GPQA DiamondEpoch GPQA Diamond · source setting 2: Models answer graduate-level science multiple-choice questions Epoch 90.6% #29 of 133
OTIS Mock AIME 20242025 OTIS Mock AIME 2024 2025: Models solve the stated OTIS mock AIME mathematics problems Epoch 100% #1 of 130
No benchmark matches
CheaperGPT-6 Astrahigh-5%-5%$1.73Monitoring can return HTTP 403+2
OpenMiMo v2.6 Prothinking-11%-93%$0.13No PDF input+4
Muse Spark 1.3max-12%-12%$1.60Only auto tool_choice is supported+5
MiniMax M3.1 Flashthinking-14%-88%$0.22No PDF input+1
Kimi K3maxOpen-16%+10%$2.00Sampling parameters are fixed+3
Step 5 PreviewthinkingOpen-16%-44%$1.03No PDF input+4
DeepSeek V4.1 Flash 1017maxOpen-22%-86%$0.26No PDF input+5