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 23.7 #91 of 337
Knowledge 6
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 40.9% #44 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 76.2% #152 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA -4.1 #79 of 321
IFBench (instructions) IFBench (instructions): Checks whether the model follows precise formatting and content instructions AA 43% #141 of 248
EQ-Bench3 EQ-Bench 3: Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking EQ-Bench 1,546 #16 of 50
SimpleQA Verifiedv1 SimpleQA Verified v1: Accuracy on short factual questions with verified answers, answered without web access 39% #43 of 83
Reasoning 1
Chess Puzzles Chess Puzzles: Models choose the best next move from a novel chess position Epoch 5% #65 of 68
Math 2
FrontierMath FrontierMath: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 20.7% #21 of 40
FrontierMath Tier4 FrontierMath Tier 4: Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 4.2% #16 of 33
Science 2
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 13.2% #154 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 0.3% #148 of 324
Coding 4
WebDev Arena WebDev Arena: Human pairwise preferences in the stated webdev vote pool Arena 1,494 #47 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 1,494 #47 of 97
SWE-bench Pro (private) SWE-bench Pro (private): Agents resolve long-horizon professional repository tasks; public and private splits are separate Scale 23.4% #7 of 9
Terminal-Bench Hard Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal AA 40.9% #31 of 242
Agents 3
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 73.4% #7 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 24.7% #18 of 21
MCP-Atlasfull 1000 MCP-Atlas · full 1000: Tool orchestration across MCP servers on the stated task split Scale 69.8% #7 of 16
Multimodal 2
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 71.2% #67 of 127
MMMUvalidation MMMU · validation: Image-grounded university-level questions across 30 disciplines; validation and test sets remain separate MMMU 80.7% #3 of 41
Writing 17
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,469 #12 of 195+14 versions & slices overall Arena Text · overall: Human pairwise preferences in the stated text vote pool Arena 1,474 #27 of 196Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool Arena 1,470 #61 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,461 #32 of 188French Arena Text · french: Human pairwise preferences in the stated text vote pool Arena 1,471 #34 of 149German Arena Text · german: Human pairwise preferences in the stated text vote pool Arena 1,449 #37 of 154instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,472 #21 of 188Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool Arena 1,412 #39 of 138Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool Arena 1,424 #28 of 144non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,438 #39 of 188Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool Arena 1,439 #37 of 133Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,447 #38 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,453 #40 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,451 #37 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 1,443 #27 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 73.1% #31 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 1,687 #32 of 91
Health & law 3
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench Vals 84.6% #27 of 89
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 49.2% #21 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 85.3% #26 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 70.7% #118 of 275
Arena Document Arena Document · overall: Human pairwise preferences in the stated document vote pool Arena 1,462 #16 of 33
Legacy · saturated or no longer reported 20
SWE-bench Pro (public) SWE-bench Pro (public): Agents resolve long-horizon professional repository tasks; public and private splits are separate 45.9% #6 of 19
AIME2025 · AA archived AIME 2025 · AA archived: AA evaluates the 30 integer-answer problems from the 2025 AIME mathematics competition AA 62.7% #77 of 140
LiveCodeBenchAA archived LiveCodeBench · AA archived: AA evaluates solutions to newly collected competitive-programming problems AA 73.8% #32 of 153
GPQA Diamond GPQA Diamond: Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 81% #111 of 315+1 versions & slices Epoch GPQA Diamond · source setting 2: Models answer graduate-level science multiple-choice questions Epoch 80.7% #68 of 133
BFCLv4 BFCL v4: Tests function calling, multi-turn interactions, search and memory in the v4 suite Berkeley 77.5% #1 of 39+1 versions & slices v4 · prompt mode BFCL v4 · BFCL v4 source function-call format · historical run: Tests function calling, multi-turn interactions, search and memory in the v4 suite Berkeley 33.5% #9 of 16
τ²-benchairline · 1.0.1 τ-bench official · 2-airline-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools Sierra 84% #1 of 6+4 versions & slices overall · 1.0.1 τ-bench official · 2-overall-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools Sierra 85.3% #2 of 6retail · 1.0.1 τ-bench official · 2-retail-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools Sierra 79.6% #3 of 6telecom · 1.0.1 τ-bench official · 2-telecom-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools Sierra 92.3% #2 of 6AA τ²-Bench: Tests service agents that use tools while coordinating with a simulated user AA 86.3% #58 of 245
OTIS Mock AIME 20242025 OTIS Mock AIME 2024 2025: Models solve the stated OTIS mock AIME mathematics problems Epoch 48.1% #94 of 130
SWE-bench VerifiedEpoch Swe Bench Verified: Agents modify real repositories to resolve the stated Verified issues Epoch 76.7% #6 of 24
SWE-rebenchall · 2025-10-01 to 2026-02-01 SWE-rebench · all · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 53.1% #1 of 1+5 versions & slices Go · 2025-10-01 to 2026-02-01 SWE-rebench · go · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 1Java · 2025-10-01 to 2026-02-01 SWE-rebench · java · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 1Python · 2025-10-01 to 2026-02-01 SWE-rebench · python · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 53.1% #1 of 1Rust · 2025-10-01 to 2026-02-01 SWE-rebench · rust · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 1TypeScript · 2025-10-01 to 2026-02-01 SWE-rebench · typescript · 2025-10-01 to 2026-02-01: Repository issue resolution over the exact fresh pull-request window, under the source agent version Nebius 0% #1 of 1
No benchmark matches
OpenKimi K2 0711unstated-24%-91%$0.50 / $2Context: 200K → 128K+2
OpenQwen3 235B A22B Instruct 2507unstated-33%-98%$0.09 / $0.55No image inputNOpenNanbeige4 3B Thinking 2511-34%——Context: 200K → 66K+1
Command AunstatedOpen-40%-56%$2.50 / $10No image input