Benchmark By Score Rank
Overall 1
AA Intelligence Index4.3.2 Artificial Analysis combines tests of knowledge, reasoning, coding and tool use into one score AA 30.4 #55 of 337
Knowledge 5
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 44.3% #36 of 321+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 61.6% #110 of 321
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 0.3 #69 of 321
IFBench (instructions) Checks whether the model follows precise formatting and content instructions AA 75.4% #23 of 248
EQ-Bench3 Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking EQ-Bench 1,559 #14 of 50
Reasoning 4
ARC-AGI-1public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 90.3% #30 of 61+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 78.7% #36 of 64
ARC-AGI-2public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 39.9% #30 of 63+1 versions & slices semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 43.3% #33 of 66
Math 22
MathArenaAIME 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #4 of 24+21 versions & slices HMMT February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 97% #3 of 24AIME 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 30Apex Answer accuracy on the named mathematics competition under the recorded effort MathArena 13.5% #16 of 34Apex Shortlist Answer accuracy on the named mathematics competition under the recorded effort MathArena 79.3% #9 of 30ArXivMath · December 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 52.2% #4 of 13ArXivMath · February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 37.5% #14 of 20ArXivMath · January 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 67.9% #7 of 18BrokenArXiv · February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 25.8% #4 of 15BRUMO 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #4 of 24CMIMC 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 95.6% #1 of 22Project Euler Answer accuracy on the named mathematics competition under the recorded effort MathArena 81.6% #4 of 11HMMT February 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #1 of 30HMMT November 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 95.8% #1 of 14Kangaroo 2025 1-2 Answer accuracy on the named mathematics competition under the recorded effort MathArena 80.2% #8 of 17Kangaroo 2025 11-12 Answer accuracy on the named mathematics competition under the recorded effort MathArena 95.8% #7 of 17Kangaroo 2025 3-4 Answer accuracy on the named mathematics competition under the recorded effort MathArena 74% #5 of 17Kangaroo 2025 5-6 Answer accuracy on the named mathematics competition under the recorded effort MathArena 80% #6 of 17Kangaroo 2025 7-8 Answer accuracy on the named mathematics competition under the recorded effort MathArena 89.2% #8 of 16Kangaroo 2025 9-10 Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 16SMT 2025 Answer accuracy on the named mathematics competition under the recorded effort MathArena 92% #2 of 24ArXivMath · January 2026 · GPT-5.2 · low effort Answer accuracy on the named mathematics competition under the recorded effort MathArena 47.9% #1 of 1
Science 2
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 37.7% #53 of 337
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 11.6% #53 of 324
Coding 4
WebDev Arena Human pairwise preferences in the stated webdev vote pool Arena 1,417 #63 of 97+1 versions & slices full cohort Human pairwise preferences in the stated webdev vote pool Arena 1,417 #63 of 97
SWE-bench Pro (private) Agents resolve long-horizon professional repository tasks; public and private splits are separate Scale 23.8% #6 of 9
Terminal-Bench Hard Agents solve difficult software, system-administration and data-processing tasks in a terminal AA 47% #16 of 242
Agents 4
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 71.8% #9 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 32.2% #12 of 21
MCP-Atlasfull 1000 Tool orchestration across MCP servers on the stated task split Scale 67.6% #9 of 16
τ³-BankingAA Agents retrieve banking knowledge and execute tools to resolve customer-support workflows AA 11.1% #88 of 134
Multimodal 2
MMMU-Pro (vision) College-level questions that require understanding images: charts, diagrams, photos AA 74.6% #54 of 127
Arena Vision Human pairwise preferences in the stated vision vote pool Arena 1,246 #47 of 88
Writing 18
Arena Textcreative writing Human pairwise preferences in the stated text vote pool Arena 1,386 #80 of 195+14 versions & slices overall Human pairwise preferences in the stated text vote pool Arena 1,437 #68 of 196Chinese Human pairwise preferences in the stated text vote pool Arena 1,447 #83 of 181English Human pairwise preferences in the stated text vote pool Arena 1,422 #80 of 188French Human pairwise preferences in the stated text vote pool Arena 1,454 #51 of 149German Human pairwise preferences in the stated text vote pool Arena 1,444 #43 of 154instruction following Human pairwise preferences in the stated text vote pool Arena 1,409 #72 of 188Japanese Human pairwise preferences in the stated text vote pool Arena 1,420 #34 of 138Korean Human pairwise preferences in the stated text vote pool Arena 1,370 #65 of 144non-English Human pairwise preferences in the stated text vote pool Arena 1,405 #74 of 188Polish Human pairwise preferences in the stated text vote pool Arena 1,423 #55 of 133Russian Human pairwise preferences in the stated text vote pool Arena 1,412 #71 of 186Spanish Human pairwise preferences in the stated text vote pool Arena 1,413 #74 of 155overall · full cohort Human pairwise preferences in the stated text vote pool Arena 1,416 #79 of 188creative writing · full cohort Human pairwise preferences in the stated text vote pool Arena 1,376 #79 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 71.6% #35 of 95
EQ-Bench Creative Writingv3 Creative Writing v3 · Elo: Judge-rated creative writing under the v3 prompt set and pairwise ranking EQ-Bench 1,703 #30 of 91
Lech Mazur Short-Story Judges compare stories written to ten required elements; the comparison score can be negative Lech 0.3 #21 of 44
Health & law 3
Vals LegalBench Legal reasoning and legal-language understanding on LegalBench Vals 82.8% #50 of 89
Vals MedCode Medical coding accuracy on patient encounters Vals 49.8% #16 of 78
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 84.4% #31 of 80
Documents 3
AA-LCR (long context)1.1 Questions that need reading and connecting facts across very long documents (around 100K tokens) AA 82.7% #22 of 275
OmniDocBenchv1.6 · full End-to-end document parsing, averaging text recognition, table structure and formula recognition under the stated full-set protocol OmniDocBench 86.6% #18 of 22
Arena Document Human pairwise preferences in the stated document vote pool Arena 1,405 #32 of 33
Legacy · saturated or no longer reported 9
SWE-bench Pro (public) Agents resolve long-horizon professional repository tasks; public and private splits are separate 29.9% #13 of 19
AIME2025 · AA archived AA evaluates the 30 integer-answer problems from the 2025 AIME mathematics competition AA 99% #1 of 140
LiveCodeBenchAA archived AA evaluates solutions to newly collected competitive-programming problems AA 89.4% #3 of 153
GPQA Diamond Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 90.3% #33 of 315
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 83% #2 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 84.8% #3 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 81.6% #2 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 89.7% #4 of 6AA Tests service agents that use tools while coordinating with a simulated user AA 84.8% #65 of 245
No benchmark matches
OpenKimi K2.6unstatedwins 57%-64%$0.95 / $4Context: 400K → 256K
CheaperMuse Spark 1.3maxwins 59%-58%$1.60Only auto tool_choice is supported+5
CheaperClaude Sonnet 5highwins 59%-17%$1.79
Grok 4.6highwins 57%-38%$1.48
MiniMax M3unstatedOpenwins 54%-89%$0.30 / $1.20No strict JSON+1
Gemma 4 31BunstatedOpenwins 53%-97%$0.10 / $0.34Context: 400K → 256K