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 9.3 #220 of 337
Knowledge 5
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 24.9% #114 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 88.9% #238 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA -41.9 #181 of 321
IFBench (instructions) IFBench (instructions): Checks whether the model follows precise formatting and content instructions AA 43% #142 of 248
EQ-Bench3 EQ-Bench 3: Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking EQ-Bench 807 #44 of 50
Reasoning 4
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 7.1% #60 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 4.4% #62 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 0% #54 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 0% #59 of 66
Science 3
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 4.9% #279 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 0% #180 of 324
SciCode SciCode: Writing code for real scientific computing problems (physics, biology, math) AA 31.7% #120 of 140
Coding 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 0% #75 of 132+1 versions & slices 2.1 Terminal-Bench 2.1 · source setting 2: Older, easier version of Terminal-Bench; Still widely reported AA 7.9% #130 of 156
Terminal-Bench Hard Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal AA 6.8% #144 of 242
Agents 4
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation AA 0.2% #128 of 133
AA-Briefcase1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating AA 0 #129 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 0 #122 of 175
τ³-BankingAA τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 3.7% #129 of 134
Multimodal 3
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 62.1% #92 of 127
MMMUvalidation MMMU · validation: Image-grounded university-level questions across 30 disciplines; validation and test sets remain separate MMMU 73.4% #12 of 41
Arena Vision Arena Vision · overall: Human pairwise preferences in the stated vision vote pool Arena 1,142 #75 of 88
Writing 17
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,307 #127 of 195+14 versions & slices overall Arena Text · overall: Human pairwise preferences in the stated text vote pool Arena 1,327 #139 of 196Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool Arena 1,277 #142 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,302 #148 of 188French Arena Text · french: Human pairwise preferences in the stated text vote pool Arena 1,259 #128 of 149German Arena Text · german: Human pairwise preferences in the stated text vote pool Arena 1,291 #112 of 154instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,267 #143 of 188Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool Arena 1,206 #109 of 138Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool Arena 1,204 #120 of 144non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,269 #143 of 188Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool Arena 1,276 #118 of 133Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,286 #136 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,293 #126 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,287 #147 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 1,266 #139 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 31.9% #86 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 860 #80 of 91
Health & law 4
MLCR-AA MLCR-AA: Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 2.2% #55 of 64
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench Vals 77.8% #72 of 89
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 36.5% #59 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 54.2% #78 of 80
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 50% #178 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 0.8% #122 of 133
Legacy · saturated or no longer reported 11
AIME2025 · AA archived AIME 2025 · AA archived: AA evaluates the 30 integer-answer problems from the 2025 AIME mathematics competition AA 19.3% #115 of 140
LiveCodeBenchAA archived LiveCodeBench · AA archived: AA evaluates solutions to newly collected competitive-programming problems AA 39.7% #106 of 153
GPQA Diamond GPQA Diamond: Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 67.1% #202 of 315
Aider Polyglottwo attempts Aider Polyglot · two attempts: Models edit code for 225 Exercism exercises in six languages, with a second attempt after test feedback Aider 15.6% #13 of 15
SWE-rebenchall · 2025-01-01 to 2025-09-01 SWE-rebench · all · 2025-01-01 to 2025-09-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 11% #2 of 4+5 versions & slices Go · 2025-01-01 to 2025-09-01 SWE-rebench · go · 2025-01-01 to 2025-09-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4Java · 2025-01-01 to 2025-09-01 SWE-rebench · java · 2025-01-01 to 2025-09-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4Python · 2025-01-01 to 2025-09-01 SWE-rebench · python · 2025-01-01 to 2025-09-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 11% #2 of 4Rust · 2025-01-01 to 2025-09-01 SWE-rebench · rust · 2025-01-01 to 2025-09-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 4TypeScript · 2025-01-01 to 2025-09-01 SWE-rebench · typescript · 2025-01-01 to 2025-09-01: 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 17.8% #203 of 245
No benchmark matches
CheaperAgnes 2.5 Pro Beta×3.8-51%$0.10 / $0.30
BetterGPT-6 Lunalow×2.3-88%$0.005Function calling only with none · Chat Completions+3
OpenDeepSeek V4.1 Flash 1017max×4.8+584%$0.26Memory (4-bit): 241 GB → 458 GB+5
CheaperClaude Haiku 5.5low×3.2-36%$0.024Remove sampling parameters+5
Ling 3.0 Flash VLthinkingOpen×2.6-70%$0.06 / $0.18Context: 1M → 262K
MiMo v2.5thinkingOpen×2.3-50%$0.019
GLM 5.3maxOpen×4.8+5,137%$2.01Memory (4-bit): 241 GB → 452 GB+5
Step 5 PreviewthinkingOpen×4.7+2,580%$1.03Memory (4-bit): 241 GB → 363 GB+3