Q1 How high is its intelligence index across knowledge work, coding and science? AA Intelligence Index v5 · Bottom 38%#35 Q1 How high is its intelligence index across knowledge work, coding and science? AA Intelligence Index v5 · Bottom 38%#35 Q2 Which AI gives the answers people prefer most? Arena Text · Top 27%#29 Q3 Can it solve problems it never trained on? LiveBench 2.0 · Bottom 22%#19 Q4 On the old v1, could it solve problems it had not seen in training? LiveBench Overall v1 · Bottom 25%#40
Q1 Bottom 38%#35 of 54 27.8 Q2 1475 Q3 60.7 Q4 71.6 About AA Intelligence Index v5 More about AA Intelligence Index v5
Q1 Can it handle customer service by the rules? τ³-Banking · Top 6%#4 Q1 Can it handle customer service by the rules? τ³-Banking · Top 6%#4 Q2 Can it finish long tasks where early decisions must hold up to the end? LiveBench 2.0 Long Agentic · Bottom 22%#19 Q3 Can it complete multi-step tasks in a real web app using a browser? LiveBench 2.0 Browser · Bottom 31%#17
Q1 Top 6%#4 of 79 47.2% Q2 20.7 Q3 59.8 About τ³-Banking More about τ³-Banking
Q1 Can it pull scattered facts from several long documents into one answer? Long Context Reasoning · Top 30%#36 Q1 Can it pull scattered facts from several long documents into one answer? Long Context Reasoning · Top 30%#36 Q2 Can it read tables and reshape them as needed? LiveBench 2.0 Data Analysis · Bottom 40%#15 Q3 Which AI gives the business answers people prefer most? Arena Business, Management & Finance · Top 32%#34 Q1 Top 30%#36 of 121 80.0% Q2 76.4 Q3 1471 Q4 52.8% Q5 57.9% Q6 31.6% Q7 45.2% Q8 10.4% Q9 73.7 Q10 76.4 Q11 60.6% Q12 14.0% Q13 6.7% About Long Context Reasoning More about Long Context Reasoning
Q1 Which AI gives answers people prefer most, without false claims? Arena Text Factuality · Top 41%#42 Q1 Which AI gives answers people prefer most, without false claims? Arena Text Factuality · Top 41%#42 Q2 Can it use search and code to solve problems that stump experts? Humanity's Last Exam · Top 37%#16 Q3 Can it research an open question and answer with a sourced report? LiveBench 2.0 Research · Bottom 18%#20 Q1 Top 41%#42 of 103 1461 Q2 55.3% Q3 73.9 Q4 2.9% Q5 0.149 About Arena Text Factuality More about Arena Text Factuality
Q1 Can AI carry out long development tasks autonomously by operating a computer itself? Terminal-Bench 4.0 · Top 24%#20 Q1 Can AI carry out long development tasks autonomously by operating a computer itself? Terminal-Bench 4.0 · Top 24%#20 Q2 Which AI do people prefer most for coding answers? Arena Coding · Top 25%#26 Q3 Can AI test and fix code itself until it completes the given task? LiveBench Agentic Coding v1 · Top 33%#17 Q1 Top 24%#20 of 84 32.8% Q2 1526 Q3 56.8 Q4 304 Q5 84.3% Q6 71.5 Q7 1611 Q8 51.6% Q9 70.9 Q10 31.8% Q11 63.4% Q12 1515 Q13 79.0 Q14 0.0% Q15 30.8% Q16 18.1% Q17 36.8% 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 42%#45 Q1 Which AI gives the writing and literature answers people prefer most? Arena Writing, Literature & Language · Top 42%#45 Q2 Which AI do people prefer most in long conversations? Arena Multi-turn · Top 34%#36 Q3 Which AI writes the creative pieces people prefer most? Arena Creative Writing · Top 50%#54 Q1 Top 42%#45 of 108 1452 Q2 1479 Q3 1435 Q4 1474 Q5 77.3 Q6 88.9% Q7 1439 Q8 52.8 About Arena Writing, Literature & Language More about Arena Writing, Literature & Language
Q Can it make accurate judgment calls across many work domains? Hanno Labs DecisionBench · Top 20%#1
Q Can it make accurate judgment calls across many work domains? Hanno Labs DecisionBench · Top 20%#1
Top 20%#1 of 5 74.1 About Hanno Labs DecisionBench More about Hanno Labs DecisionBench
Q1 Can it solve problems that stump experts across fields? Humanity's Last Exam · Top 35%#42 Q1 Can it solve problems that stump experts across fields? Humanity's Last Exam · Top 35%#42 Q2 Can it solve research-level physics problems? CritPt · Top 44%#38 Q3 Which AI gives the math answers people prefer most? Arena Math · Top 8%#8 Q1 Top 35%#42 of 122 39.9% Q2 15.4% Q3 1509 Q4 91.2% Q5 93.9% Q6 81.2 Q7 55.8% Q8 17.1% Q9 21.0% Q10 52.5% Q11 1493 Q12 1495 Q13 77.6 Q14 77.6 Q15 91.0% Q16 65.8% Q17 21.0% Q18 8.0% 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 · Top 23%#15 Q1 Which AI gives the photo and chart answers people prefer most? Arena Vision · Top 23%#15 Q2 Can it read charts and answer accurately? CharXiv · Top 17%#5
Q1 Top 23%#15 of 66 1297 Q2 89.4% About Arena Vision More about Arena Vision
Q1 Which AI do people prefer most when asking in Korean? Arena Korean · Top 36%#32 Q1 Which AI do people prefer most when asking in Korean? Arena Korean · Top 36%#32 Q2 Which AI do people prefer most when asking in Japanese? Arena Japanese · Top 33%#29 Q3 Which AI do people prefer most on tough questions? Arena Hard Prompts · Top 26%#28 Q1 Top 36%#32 of 91 1430 Q2 1457 Q3 1501 Q4 1513 Q5 1490 Q6 1476 Q7 56.1% Q8 98.8% Q9 84.0% Q10 57.7 Q11 63.4% Q12 82.6% Q13 87.5% Q14 80.4% Q15 78.4% About Arena Korean More about Arena Korean
Q1 What would it score on Korea's 2026 CSAT (Suneung)? CSAT 2026 · Bottom 46%#34 Q1 What would it score on Korea's 2026 CSAT (Suneung)? CSAT 2026 · Bottom 46%#34 Q2 What would it score on Korea's 2027 CSAT (Suneung)? CSAT 2027 · Bottom 42%#22
Q1 Bottom 46%#34 of 61 376 Reasoning effort: High Q2 396 About CSAT 2026 More about CSAT 2026