Q1 How strong is it across many tests combined? AA Intelligence Index · Bottom 23%#91 Q1 How strong is it across many tests combined? AA Intelligence Index · Bottom 23%#91 Q2 Which AI gives the answers people prefer most? Arena Text · Bottom 40%#60 Q3 Does it have broad knowledge across expert fields? MMLU-Pro · Top 25%#7
Q1 #91 of 116 16.1 Q2 1446 Q3 86.2% About AA Intelligence Index More about AA Intelligence Index
Q1 Can it handle customer service by the rules? τ³-Banking · Bottom 17%#67 Q1 Can it handle customer service by the rules? τ³-Banking · Bottom 17%#67 Q2 Can it run a small business alone for a year and make money? Vending-Bench 2 · Bottom 15%#41 Q3 Can it handle customer service by the rules? TAU2 · Bottom 17%#67 Q1 #67 of 79 9.7% Q2 $574 Q3 54.1% Q4 8.8% Q5 11.4% About τ³-Banking More about τ³-Banking
Q1 Can it pull scattered facts from several long documents into one answer? Long Context Reasoning · Bottom 24%#89 Q1 Can it pull scattered facts from several long documents into one answer? Long Context Reasoning · Bottom 24%#89 Q2 Which AI gives the business answers people prefer most? Arena Business, Management & Finance · Bottom 31%#68 Q3 Can its work match what professionals produce? GDPval · Bottom 20%#5 Q1 #89 of 115 69.0% Q2 1437 Q3 23.3% Q4 60.8% Q5 50.6% Q6 1421 Q7 41.6% About Long Context Reasoning More about Long Context Reasoning
Q1 Which AI gives answers people prefer most, without false claims? Arena Text Factuality · Top 38%#34 Q1 Which AI gives answers people prefer most, without false claims? Arena Text Factuality · Top 38%#34 Q2 Can it carry out multi-step web research? Deep Research Bench · Bottom 24%#14 Q3 Can it answer questions that need many sources, without missing anything? DeepSearchQA · Bottom 29%#11 Q1 #34 of 90 1462 Q2 41.5% Q3 24.6% Q4 7.0% Q5 93.0% Q6 60.1 About Arena Text Factuality More about Arena Text Factuality
Q1 Can it operate a computer itself to finish hours-long development jobs? Terminal-Bench 4.0 · Bottom 27%#58 Q1 Can it operate a computer itself to finish hours-long development jobs? Terminal-Bench 4.0 · Bottom 27%#58 Q2 Which AI gives the coding answers people prefer most? Arena Coding · Bottom 21%#78 Q3 Can it find better answers to optimization problems with no single right answer? ALE-Bench · Bottom 37%#48 Q1 #58 of 78 0.0% Q2 1465 Q3 786 Q4 28.5% Q5 26.5% Q6 33.3 Q7 1227 Q8 80.1% Q9 46.3% Q10 57.6% Q11 54.0% Q12 3.9% Q13 1460 Q14 32.6% Q15 1.51× About Terminal-Bench 4.0 More about Terminal-Bench 4.0
Q1 Which AI gives the writing and literature answers people prefer most? Arena Writing, Literature & Language · Top 48%#46 Q1 Which AI gives the writing and literature answers people prefer most? Arena Writing, Literature & Language · Top 48%#46 Q2 Which AI do people prefer most in long conversations? Arena Multi-turn · Bottom 38%#62 Q3 Does it follow tricky rules like format and length without missing any? IFBench · Bottom 13%#72 Q1 #46 of 97 1445 Q2 1448 Q3 48.7% Q4 1443 Q5 1438 Q6 8.38 / 10 Q7 73.6% Q8 1431 About Arena Writing, Literature & Language More about Arena Writing, Literature & Language
Q1 Can it solve problems that stump experts across fields? Humanity's Last Exam · Bottom 34%#78 Q1 Can it solve problems that stump experts across fields? Humanity's Last Exam · Bottom 34%#78 Q2 Can it solve research-level physics problems? CritPt · Bottom 24%#61 Q3 Which AI gives the math answers people prefer most? Arena Math · Bottom 39%#59 Q1 #78 of 116 22.5% Q2 2.0% Q3 1439 Q4 84.4% Q5 84.7% Q6 87.7 Q7 87.7% Q8 24.6% Q9 0.0% Q10 1472 Q11 1447 Q12 96.7% Q13 14.1% Q14 4.2% Q15 41.0% Q16 4.9% Q17 20.0% Q18 82.4% Q19 34.8 About Humanity's Last Exam More about Humanity's Last Exam
Q1 Which AI gives the photo and chart answers people prefer most? Arena Vision · Bottom 37%#38 Q1 Which AI gives the photo and chart answers people prefer most? Arena Vision · Bottom 37%#38 Q2 Can it read charts and answer accurately? CharXiv · Bottom 10%#29 Q3 Can it rotate and fold objects in its head? SpatialViz-Bench · Top 25%#1 Q4 Can it solve college-level problems with images and diagrams? MMMU-Pro · Bottom 19%#31
Q1 #38 of 58 1263 Q2 69.6% Q3 44.7% Q4 68.0% About Arena Vision More about Arena Vision
Q1 Which AI do people prefer most when asking in Korean? Arena Korean · Bottom 50%#40 Q1 Which AI do people prefer most when asking in Korean? Arena Korean · Bottom 50%#40 Q2 Which AI do people prefer most when asking in Japanese? Arena Japanese · Top 34%#25 Q3 Can it use expert knowledge in many languages? MMMLU · Top 39%#7 Q1 #40 of 79 1414 Q2 1445 Q3 89.5% Q4 1458 Q5 1458 Q6 1469 Q7 1469 Q8 90.0% Q9 58.1% Q10 74.3% Q11 46.9% Q12 63.2% Q13 61.5% Q14 93.2% Q15 93.8% Q16 92.8% Q17 82.5% Q18 80.0% Q19 31.6% Q20 95.7% Q21 84.7% Q22 0.5% Q23 77.2% Q24 15.8% Q25 78.2% Q26 85.9% Q27 75.3% Q28 45.8% Q29 84.9% Q30 24.4% Q31 5.0% Q32 26.0% About Arena Korean More about Arena Korean
Nano Banana 2.1