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 33.8 #45 of 334
Knowledge 20
Arena Text Human pairwise preferences in the stated text vote pool Arena 1,481 #19 of 196+12 versions & slices English Human pairwise preferences in the stated text vote pool Arena 1,470 #27 of 188instruction following Human pairwise preferences in the stated text vote pool Arena 1,478 #14 of 188non-English Human pairwise preferences in the stated text vote pool Arena 1,467 #17 of 188overall · full cohort Human pairwise preferences in the stated text vote pool Arena 1,471 #20 of 188Russian Human pairwise preferences in the stated text vote pool Arena 1,473 #21 of 186Chinese Human pairwise preferences in the stated text vote pool Arena 1,533 #10 of 181Spanish Human pairwise preferences in the stated text vote pool Arena 1,468 #22 of 155German Human pairwise preferences in the stated text vote pool Arena 1,480 #15 of 154French Human pairwise preferences in the stated text vote pool Arena 1,485 #22 of 149Korean Human pairwise preferences in the stated text vote pool Arena 1,459 #11 of 144Japanese Human pairwise preferences in the stated text vote pool Arena 1,498 #5 of 138Polish Human pairwise preferences in the stated text vote pool Arena 1,464 #19 of 133
AA Omniscience Index Scores factual knowledge while penalizing incorrect answers to encourage admitting uncertainty AA 18.1 #27 of 318
SimpleQA Verifiedv1 Accuracy on short factual questions with verified answers, answered without web access 63% #14 of 83
AA OmniscienceAccuracy Measures correct answers on the AA Omniscience factual knowledge test AA 56.8% #8 of 318+1 versions & slices Hallucination Rate Measures how often attempted answers are incorrect on the AA Omniscience factual knowledge test AA 89.6% #242 of 318
IFBench (instructions) Checks whether the model follows precise formatting and content instructions AA 71% #46 of 245
Mystery Game Puzzles Models choose the best move in puzzle positions from an undisclosed game Epoch 56% #9 of 64
EQ-Bench3 Judge-rated responses to social and emotionally nuanced scenarios, scored through the source pairwise ranking EQ-Bench 1,577 #7 of 50
Reasoning 6
Chess Puzzles Models choose the best next move from a novel chess position Epoch 63% #4 of 68
ARC-AGI-2semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 70.4% #19 of 66+1 versions & slices public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 70.7% #17 of 63
ARC-AGI-1semi-private Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 92.2% #20 of 64+1 versions & slices public evaluation Models infer transformations from examples and solve unseen grid puzzles; the evaluation split is explicit ARC 96.5% #17 of 61
ARC-AGI-3semi-private Measures interactive game solving through relative human action efficiency ARC 0.4% #13 of 18
Math 26
FrontierMath Tiers 1–3 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 85.3% #13 of 69
FrontierMath Tier 4 (v2) Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 72.5% #12 of 54
MathArenaApex Answer accuracy on the named mathematics competition under the recorded effort MathArena 80.2% #2 of 34+22 versions & slices Apex Shortlist Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.4% #1 of 30AIME 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 24HMMT February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.5% #1 of 24ArXivMath · June 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 83.6% #7 of 24BrokenArXiv · June 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 69.4% #7 of 24ArXivMath · February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 73.4% #2 of 20ArXivMath · May 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 77.5% #7 of 19ArXivMath · January 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 73.9% #2 of 18BrokenArXiv · May 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 50% #6 of 18ArXivMath · April 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 68.8% #4 of 17Kangaroo 2025 1-2 Answer accuracy on the named mathematics competition under the recorded effort MathArena 95.8% #1 of 17Kangaroo 2025 11-12 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.3% #2 of 17Kangaroo 2025 3-4 Answer accuracy on the named mathematics competition under the recorded effort MathArena 89.6% #1 of 17Kangaroo 2025 5-6 Answer accuracy on the named mathematics competition under the recorded effort MathArena 90% #1 of 17Kangaroo 2025 7-8 Answer accuracy on the named mathematics competition under the recorded effort MathArena 95.8% #3 of 16Kangaroo 2025 9-10 Answer accuracy on the named mathematics competition under the recorded effort MathArena 100% #1 of 16BrokenArXiv · April 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 72.1% #2 of 15BrokenArXiv · February 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 68.2% #1 of 15ArXivMath · March 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 77.5% #1 of 14BrokenArXiv · March 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 73.7% #1 of 13USAMO 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 98.2% #1 of 8ArXivLean · March 2026 Answer accuracy on the named mathematics competition under the recorded effort MathArena 17.1% #1 of 5
FrontierMath Erdos Models solve expert-written advanced mathematics problems on the named tier and edition Epoch 0% #6 of 13
Science 2
Humanity's Last Exam Very hard expert questions across many fields, written to stump today’s models AA 42.4% #35 of 334
CritPt (physics) Research-level physics problems that need many careful reasoning steps AA 18.6% #32 of 321
Health & law 5
Vals LegalBench Legal reasoning and legal-language understanding on LegalBench Vals 86.5% #10 of 89
Vals MedScribe Accuracy of clinical notes produced from consultation transcripts Vals 86.9% #19 of 80
Vals MedCode Medical coding accuracy on patient encounters Vals 49.1% #23 of 78
Vals Legal Research Bench Grounded legal research; a task passes only if all criteria pass Vals 40.4% #24 of 69
MLCR-AA Models synthesize long medical records; the headline credits concise answers judged both accurate and complete AA 17.8% #26 of 64
Documents 2
AA-LCR (long context)1.1 Questions that need reading and connecting facts across very long documents (around 100K tokens) AA 83% #16 of 273
GDP.pdfAA all-pass Models reason over professional PDF documents; an attempt passes only when every atomic criterion passes 20.6% #24 of 133
Legacy · saturated or no longer reported 3
GPQA Diamond Graduate-level biology, chemistry and physics questions that can’t be solved by a quick web search 92.6% #16 of 312+1 versions & slices Epoch Models answer graduate-level science multiple-choice questions Epoch 90.7% #28 of 133
OTIS Mock AIME 20242025 Models solve the stated OTIS mock AIME mathematics problems Epoch 84.4% #62 of 130
No benchmark matches
OpenGLM 5.3 Flashmax-5%-69%$0.28Context: 1.05M → 1M+5
CheaperGemini 4 Argonmedium+30%-64%*$2 / $10Context: 1.05M → 1M+3
DeepSeek V4 Pro 0813highOpen-10%-82%$1.32 / $3.96Context: 1.05M → 1M+6
Kimi K3maxOpen-13%+122%$2.00Context: 1.05M → 1M+4TInkling SmallxhighOpen-44%-94%$0.45 / $1.20Context: 1.05M → 256K+1
Qwen3.8 27BxhighOpen-45%+12%$1.01Context: 1.05M → 262K+5