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 29.7 #58 of 337
Knowledge 8 Leads on 1
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 54.9% #10 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 50.9% #91 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 31.9 #10 of 321
IFBench (instructions) IFBench (instructions): Checks whether the model follows precise formatting and content instructions AA 77.1% #12 of 248
EQ-Bench3 EQ-Bench 3: Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking EQ-Bench 1,538 #18 of 50
SimpleQA Verifiedv1 SimpleQA Verified v1: Accuracy on short factual questions with verified answers, answered without web access 77.5% #1 of 83
Mystery Game Puzzles Mystery Game Puzzles: Models choose the best move in puzzle positions from an undisclosed game Epoch 34% #19 of 64+1 versions & slices August 2026 run Mystery Game Puzzles · Epoch independent run · historical run: Models choose the best move in puzzle positions from an undisclosed game Epoch 29% #1 of 1
Reasoning 6
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 97.2% #16 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% #7 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 88.1% #15 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 77.1% #16 of 66
ARC-AGI-3semi-private ARC-AGI-3 · semi-private: Measures interactive game solving through relative human action efficiency ARC 0.4% #14 of 18
Chess Puzzles Chess Puzzles: Models choose the best next move from a novel chess position Epoch 51% #14 of 68
Math 31
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings AA 41.3% #19 of 43
MathArenaAIME 2026 MathArena · AIME 2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #4 of 24+25 versions & slices HMMT February 2026 MathArena · HMMT February 2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 94.7% #7 of 24Apex MathArena · apex · Apex: Answer accuracy on the named mathematics competition under the recorded effort MathArena 60.9% #5 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 63.8% #7 of 17ArXivMath · December 2025 MathArena · arxiv · 12/2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 66.2% #1 of 13ArXivMath · February 2026 MathArena · arxiv · 02/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 62.5% #3 of 20ArXivMath · January 2026 MathArena · arxiv · 01/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 70.7% #6 of 18ArXivMath · June 2026 MathArena · arxiv · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 66.7% #16 of 24ArXivMath · March 2026 MathArena · arxiv · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 68.3% #3 of 14ArXivMath · May 2026 MathArena · arxiv · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 51.7% #17 of 19BrokenArXiv · April 2026 MathArena · arxiv_false · 04/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 25.8% #8 of 15BrokenArXiv · February 2026 MathArena · arxiv_false · 02/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 18.6% #5 of 15BrokenArXiv · June 2026 MathArena · arxiv_false · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 17.6% #19 of 24BrokenArXiv · March 2026 MathArena · arxiv_false · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 13.8% #7 of 13BrokenArXiv · May 2026 MathArena · arxiv_false · 05/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 17% #14 of 18ArXivLean · June 2026 MathArena · arxivlean · 06/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 10.9% #7 of 9ArXivLean · March 2026 MathArena · arxivlean · 03/2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 14.6% #3 of 5Project Euler MathArena · euler · Project Euler: Answer accuracy on the named mathematics competition under the recorded effort MathArena 89% #1 of 11Kangaroo 2025 1-2 MathArena · kangaroo · Kangaroo 2025 1-2: Answer accuracy on the named mathematics competition under the recorded effort MathArena 86.5% #5 of 17Kangaroo 2025 11-12 MathArena · kangaroo · Kangaroo 2025 11-12: Answer accuracy on the named mathematics competition under the recorded effort MathArena 97.5% #5 of 17Kangaroo 2025 3-4 MathArena · kangaroo · Kangaroo 2025 3-4: Answer accuracy on the named mathematics competition under the recorded effort MathArena 76% #4 of 17Kangaroo 2025 5-6 MathArena · kangaroo · Kangaroo 2025 5-6: Answer accuracy on the named mathematics competition under the recorded effort MathArena 86.7% #3 of 17Kangaroo 2025 7-8 MathArena · kangaroo · Kangaroo 2025 7-8: Answer accuracy on the named mathematics competition under the recorded effort MathArena 90% #6 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 16USAMO 2026 MathArena · usamo · USAMO 2026: Answer accuracy on the named mathematics competition under the recorded effort MathArena 74.4% #3 of 8
FrontierMath FrontierMath: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 36.9% #9 of 40
FrontierMath Tier4 FrontierMath Tier 4: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 16.7% #6 of 33
FrontierMath Tier 4 (v2) FrontierMath Tier 4 (v2): Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 26.8% #32 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 59.6% #34 of 69
Science 4
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 47% #18 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 17.7% #37 of 324
Vals Science Index Vals Science Index: Scientific problem solving across the index’s component evaluations Vals 27.4% #17 of 18
SciCode SciCode: Writing code for real scientific computing problems (physics, biology, math) AA 58.7% #14 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 4% #45 of 132+1 versions & slices 2.1 Terminal-Bench 2.1 · source setting 2: Older, easier version of Terminal-Bench; Still widely reported AA 73.8% #32 of 156
DeepSWE1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition 11.7% #35 of 44+1 versions & slices 1 DeepSWE 1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition Datacurve 9.7% #15 of 20
ITBench SRE ITBench SRE: IT incident tasks: diagnose and fix broken systems like a site-reliability engineer AA 30.3% #29 of 37
WebDev Arena WebDev Arena: Human pairwise preferences in the stated webdev vote pool Arena 1,447 #53 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 1,447 #53 of 97
MirrorCode MirrorCode: Agents reimplement software projects by observing their behavior Epoch 8.9% #9 of 9
Terminal-Bench Hard Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal AA 53.8% #8 of 242
Agents 11
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 35.4% #48 of 133
AA-Briefcase1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 457 #99 of 140
APEX-Agents APEX-Agents: Agents complete long-horizon professional-services tasks across realistic applications AA 32% #10 of 29
EnterpriseOps-Gym-AA EnterpriseOps-Gym-AA: Agents complete stateful enterprise workflows in eight domains; final database state determines success AA 42.2% #19 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 14.6 #100 of 175
MCP-Atlas Public MCPAtlas Public: MCPAtlas Public as reported in this launch evaluation; the post is the authority for the recorded task and grading scope 80.6% #3 of 16
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 26% #15 of 21
EBR Bench EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs Epoch 14.3% #18 of 24
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step Epoch 26.7% #25 of 31
MCP-Atlasfull 1000 MCP-Atlas · full 1000: Tool orchestration across MCP servers on the stated task split Scale 78.2% #3 of 16
τ³-BankingAA τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 21.4% #51 of 134
Multimodal 2
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 82.4% #10 of 127
Arena Vision Arena Vision · overall: Human pairwise preferences in the stated vision vote pool Arena 1,296 #14 of 88
Writing 18
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,481 #9 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 #19 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,478 #18 of 188French Arena Text · french: Human pairwise preferences in the stated text vote pool Arena 1,486 #20 of 149German Arena Text · german: Human pairwise preferences in the stated text vote pool Arena 1,491 #10 of 154instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,466 #26 of 188Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool Arena 1,493 #6 of 138Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool Arena 1,454 #14 of 144non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,476 #12 of 188Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool Arena 1,497 #4 of 133Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,497 #8 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,480 #11 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,480 #13 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 1,482 #10 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 68.2% #42 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 1,491 #51 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 -2.2 #36 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 0% #18 of 23
MLCR-AA MLCR-AA: Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 15.6% #33 of 64
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench Vals 87.4% #4 of 89
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 59.1% #2 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 76.1% #59 of 80
Vals Legal Research Bench Vals Legal Research Bench: Grounded legal research; a task passes only if all criteria pass Vals 20.7% #48 of 69
Documents 3
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% #27 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 17.8% #30 of 133
Arena Document Arena Document · overall: Human pairwise preferences in the stated document vote pool Arena 1,444 #25 of 33
Other 1
AA Speech to textRecorded audio word error rate AA Speech-to-Text · Non-streaming WER (version unstated): Measures transcription errors as a share of reference words in the non-streaming test AA 2.8% #10 of 55
Legacy · saturated or no longer reported 10
GPQA Diamond GPQA Diamond: Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 94.1% #4 of 315+1 versions & slices Epoch GPQA Diamond · source setting 2: Models answer graduate-level science multiple-choice questions Epoch 94.4% #7 of 133
OTIS Mock AIME 20242025 OTIS Mock AIME 2024 2025: Models solve the stated OTIS mock AIME mathematics problems Epoch 95.6% #31 of 130
SWE-rebenchall · 2026-02-01 to 2026-05-15 SWE-rebench · all · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 54.9% #2 of 4+5 versions & slices Go · 2026-02-01 to 2026-05-15 SWE-rebench · go · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4Java · 2026-02-01 to 2026-05-15 SWE-rebench · java · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4Python · 2026-02-01 to 2026-05-15 SWE-rebench · python · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 54.9% #2 of 4Rust · 2026-02-01 to 2026-05-15 SWE-rebench · rust · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4TypeScript · 2026-02-01 to 2026-05-15 SWE-rebench · typescript · 2026-02-01 to 2026-05-15: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4
τ²-benchAA τ²-Bench: Tests service agents that use tools while coordinating with a simulated user AA 95.6% #16 of 245
No benchmark matches
CheaperGPT-6.1 Solmax+74%-44%$0.72Remove unsupported parameters+5
OpenMiMo v2.6 Prothinking+56%-90%$0.13No PDF input+4
CheaperClaude Sonnet 5.5high+57%-32%$0.88No forced tool use+5
DeepSeek V4.1 Flash 1017maxOpen+52%-80%$0.26No PDF input+5
GLM 5.3maxOpen+51%+54%$2.01No image input+5
Step 5 PreviewthinkingOpen+47%-21%$1.03No PDF input+4
Kimi K3maxOpen+47%+54%$2.00Sampling parameters are fixed+3