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 12.5 #175 of 337
Knowledge 4
AA OmniscienceAccuracy AA Omniscience · Accuracy: Measures correct answers on the AA Omniscience factual knowledge test AA 17.6% #192 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 52.6% #96 of 321
AA Omniscience Index AA Omniscience Index: Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA -25.8 #137 of 321
IFBench (instructions) IFBench (instructions): Checks whether the model follows precise formatting and content instructions AA 65.9% #70 of 248
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 29.7% #52 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 20.7% #54 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 0.3% #52 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 2.6% #55 of 66
Math 6
MathArenaAIME 2025 MathArena · aime · AIME 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 85% #18 of 30+5 versions & slices BRUMO 2025 MathArena · brumo · BRUMO 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 80.8% #23 of 24CMIMC 2025 MathArena · cmimc · CMIMC 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 74.4% #13 of 22HMMT February 2025 MathArena · hmmt · HMMT Feb 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 74.2% #19 of 30HMMT November 2025 MathArena · hmmt · HMMT Nov 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 81.7% #11 of 14SMT 2025 MathArena · smt · SMT 2025: Answer accuracy on the named mathematics competition under the recorded effort MathArena 85.4% #13 of 24
Science 2
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models AA 8.7% #208 of 337
CritPt (physics) CritPt (physics): Research-level physics problems that need many careful reasoning steps AA 0% #180 of 324
Coding 1
Terminal-Bench Hard Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal AA 17.4% #100 of 242
Multimodal 2
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos AA 58.2% #100 of 127
Arena Vision Arena Vision · overall: Human pairwise preferences in the stated vision vote pool Arena 1,159 #72 of 88
Writing 14
Arena Textcreative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool Arena 1,246 #161 of 195+13 versions & slices overall Arena Text · overall: Human pairwise preferences in the stated text vote pool Arena 1,338 #135 of 196Chinese Arena Text · chinese: Human pairwise preferences in the stated text vote pool Arena 1,356 #126 of 181English Arena Text · english: Human pairwise preferences in the stated text vote pool Arena 1,322 #139 of 188German Arena Text · german: Human pairwise preferences in the stated text vote pool Arena 1,327 #106 of 154instruction following Arena Text · instruction following: Human pairwise preferences in the stated text vote pool Arena 1,306 #129 of 188Japanese Arena Text · japanese: Human pairwise preferences in the stated text vote pool Arena 1,226 #102 of 138Korean Arena Text · korean: Human pairwise preferences in the stated text vote pool Arena 1,269 #103 of 144non-English Arena Text · non english: Human pairwise preferences in the stated text vote pool Arena 1,313 #130 of 188Polish Arena Text · polish: Human pairwise preferences in the stated text vote pool Arena 1,314 #104 of 133Russian Arena Text · russian: Human pairwise preferences in the stated text vote pool Arena 1,296 #132 of 186Spanish Arena Text · spanish: Human pairwise preferences in the stated text vote pool Arena 1,360 #105 of 155overall · full cohort Arena Text · overall · snapshot 2026-10-15 · native board settings: Human pairwise preferences in the stated text vote pool Arena 1,320 #134 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 1,249 #147 of 187
Health & law 3
Vals LegalBench Vals LegalBench: Legal reasoning and legal-language understanding on LegalBench Vals 50.1% #88 of 89
Vals MedCode Vals MedCode: Medical coding accuracy on patient encounters Vals 30.4% #72 of 78
Vals MedScribe Vals MedScribe: Accuracy of clinical notes produced from consultation transcripts Vals 72.9% #64 of 80
Documents 1
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 43.7% #189 of 275
Legacy · saturated or no longer reported 4
τ²-benchAA τ²-Bench: Tests service agents that use tools while coordinating with a simulated user AA 30.4% #149 of 245
AIME2025 · AA archived AIME 2025 · AA archived: AA evaluates the 30 integer-answer problems from the 2025 AIME mathematics competition AA 78.3% #48 of 140
LiveCodeBenchAA archived LiveCodeBench · AA archived: AA evaluates solutions to newly collected competitive-programming problems AA 76.3% #29 of 153
GPQA Diamond GPQA Diamond: Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 67% #203 of 315
No benchmark matches
OpenKimi K2 Thinkingunstatedsame+520%$0.47 / $2Context: 400K → 256K+1
OpenGLM 5unstatedsame+1,027%$1 / $3.20Context: 400K → 203K+1
DeepSeek V3.2 SpecialeunstatedOpensame——Context: 400K → 128K+1