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 30.1 #56 of 337
Knowledge 6 Leads on 1
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 45.8% #33 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 77.1% #154 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 3.9 #55 of 321
SimpleQA Verifiedv1 SimpleQA Verified v1: Accuracy on short factual questions with verified answers, answered without web access 50.6% #27 of 83
Mystery Game Puzzles Mystery Game Puzzles: Models choose the best move in puzzle positions from an undisclosed game Epoch 12% #50 of 64
EQ-Bench4 Active emotional and social intelligence in multi-turn conversations with 120 simulated personas EQ-Bench 1,339 #1 of 1
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 75.8% #38 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 65.7% #41 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 16.1% #38 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 12.4% #43 of 66
Chess Puzzles Chess Puzzles: Models choose the best next move from a novel chess position Epoch 18% #55 of 68
Math 22
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings AA 38.8% #21 of 43
MathArenaAIME 2026 MathArena · AIME 2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 96.7% #6 of 24+17 versions & slices HMMT February 2026 MathArena · HMMT February 2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 97% #3 of 24Apex MathArena · apex · Apex: Answer accuracy on the named mathematics competition under the recorded effort MathArena 65.6% #4 of 34Apex Shortlist MathArena · apex · Apex Shortlist: Answer accuracy on the named mathematics competition under the recorded effort MathArena 92% #2 of 30ArXivMath · April 2026 MathArena · arxiv · 04/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 54.2% #10 of 17ArXivMath · August 2026 MathArena · arxiv · 08/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 38.6% #10 of 12ArXivMath · June 2026 MathArena · arxiv · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 71.5% #11 of 24ArXivMath · May 2026 MathArena · arxiv · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 61.7% #10 of 19BrokenArXiv · April 2026 MathArena · arxiv_false · 04/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 59.4% #3 of 15BrokenArXiv · August 2026 MathArena · arxiv_false · 08/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 61.9% #6 of 12BrokenArXiv · June 2026 MathArena · arxiv_false · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 51.9% #11 of 24BrokenArXiv · May 2026 MathArena · arxiv_false · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 42.5% #8 of 18Kangaroo 2025 1-2 MathArena · kangaroo · Kangaroo 2025 1-2: Answer accuracy on the named mathematics competition under the recorded effort MathArena 93.8% #3 of 17Kangaroo 2025 11-12 MathArena · kangaroo · Kangaroo 2025 11-12: Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #2 of 17Kangaroo 2025 3-4 MathArena · kangaroo · Kangaroo 2025 3-4: Answer accuracy on the named mathematics competition under the recorded effort MathArena 77.1% #3 of 17Kangaroo 2025 5-6 MathArena · kangaroo · Kangaroo 2025 5-6: Answer accuracy on the named mathematics competition under the recorded effort MathArena 81.7% #5 of 17Kangaroo 2025 7-8 MathArena · kangaroo · Kangaroo 2025 7-8: Answer accuracy on the named mathematics competition under the recorded effort MathArena 96.7% #1 of 16Kangaroo 2025 9-10 MathArena · kangaroo · Kangaroo 2025 9-10: Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 16
FrontierMath Erdos FrontierMath Erdos: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 0% #6 of 13
FrontierMath Tier 4 (v2) FrontierMath Tier 4 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 39% #22 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 72.2% #22 of 69
Science 3
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 25% #100 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 3.1% #91 of 324
SciCode SciCode: Writing code for real scientific computing problems (physics, biology, math) AA 52.7% #38 of 140
Coding 9
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 12.6% #34 of 132+1 versions & slices 2.1 Terminal-Bench 2.1 · source setting 2: Older, easier version of Terminal-Bench; Still widely reported AA 82.4% #17 of 156
DeepSWE1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition 68.5% #12 of 44
FrontierCode1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements AA 43.6% #9 of 28
ITBench SRE ITBench SRE: IT incident tasks: diagnose and fix broken systems like a site-reliability engineer AA 47.7% #9 of 37
WebDev Arena WebDev Arena: Human pairwise preferences in the stated webdev vote pool Arena 1,655 #13 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 1,655 #13 of 97
SWE-bench Pro (v2 full) SWE-bench Pro (v2 full): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate Scale 97.7% #4 of 16
SWE-bench Pro (v2 hard) SWE-bench Pro (v2 hard): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate Scale 88.2% #6 of 16
Agents 9
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 49.3% #36 of 133
AA-Briefcase1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 986 #56 of 140
APEX-Agents APEX-Agents: Agents complete long-horizon professional-services tasks across realistic applications AA 41.3% #2 of 29
EnterpriseOps-Gym-AA EnterpriseOps-Gym-AA: Agents complete stateful enterprise workflows in eight domains; final database state determines success AA 45.3% #10 of 43
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 31.3 #58 of 175
Toolathlon-Verified Toolathlon-Verified: Toolathlon-Verified as reported in this launch evaluation; the post is the authority for the recorded task and grading scope 76.5% #2 of 6
τ³-benchbanking · 1.0.1 τ-bench official · 3-banking-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools Sierra 37.1% #10 of 21
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step Epoch 34.2% #21 of 31
τ³-BankingAA τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 41.6% #15 of 134
Multimodal 2
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 78.5% #30 of 127
Chartography Chartography: Expert chart reasoning over professional scientific, medical, financial and engineering charts Surge 26.6% #18 of 37
Writing 18
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,461 #17 of 195+14 versions & slices overall Arena Text · overall: Human pairwise preferences in the stated text vote pool Arena 1,487 #14 of 196Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool Arena 1,528 #16 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,481 #15 of 188French Arena Text · french: Human pairwise preferences in the stated text vote pool Arena 1,488 #19 of 149German Arena Text · german: Human pairwise preferences in the stated text vote pool Arena 1,489 #11 of 154instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,483 #12 of 188Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool Arena 1,489 #8 of 138Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool Arena 1,460 #10 of 144non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,464 #19 of 188Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool Arena 1,460 #22 of 133Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,481 #16 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,471 #17 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,475 #17 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 1,456 #19 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 79.6% #16 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 2,082 #5 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 2.4 #9 of 44
Health & law 6
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 5.3% #8 of 23
MLCR-AA MLCR-AA: Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 38.3% #12 of 64
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench Vals 86.2% #12 of 89
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 49.4% #19 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 88.1% #14 of 80
Vals Legal Research Bench Vals Legal Research Bench: Grounded legal research; a task passes only if all criteria pass Vals 46.2% #13 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 79.3% #52 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 18.6% #27 of 133
Legacy · saturated or no longer reported 3
GPQA Diamond GPQA Diamond: Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 84.2% #88 of 315+1 versions & slices Epoch GPQA Diamond · source setting 2: Models answer graduate-level science multiple-choice questions Epoch 84.8% #56 of 133
OTIS Mock AIME 20242025 OTIS Mock AIME 2024 2025: Models solve the stated OTIS mock AIME mathematics problems Epoch 68.9% #77 of 130
No benchmark matches
CheaperGPT-6.1 Solmax+72%-37%$0.72Remove unsupported parameters+5
OpenMiMo v2.6 Prothinking+54%-88%$0.13Pass back reasoning_content+3
CheaperClaude Sonnet 5.5high+55%-23%$0.88No forced tool use+5
DeepSeek V4.1 Flash 1017maxOpen+50%-77%$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen+49%+75%$2.01No image input+4
Step 5 PreviewthinkingOpen+45%-10%$1.03No web search+3