Every AI leaderboard

Leaderboards28Models ranked986

28 leaderboards · 986 models ranked

LiveBench leaderboard: who is #1

LiveBench · Global average, higher is better · 64 models · published Jun 25, 2026

LiveBench

Top 25 of 64 models, best variant per model as published

#ModelScoreCorrect
1Claude Fable 5.1as “claude-fable-5-1-max-effort”83.4%
2Claude Opus 5.5as “claude-opus-5-5-max-effort”83.2%
3Claude Fable 5as “claude-fable-5-max-effort”83.0%
4GPT-6 Astraas “gpt-6-astra-max”82.2%
5New, entered at #5 since Oct 2, 2026GPT-6.1 Solas “gpt-6.1-sol-max”81.6%
61down from #5 since Oct 2, 2026Muse Spark 1.3as “muse-spark-1.3-xhigh”81.6%
7New, entered at #7 since Oct 13, 2026Muse Wonder 1.0 Max Effortas “muse-wonder-1.0-max”81.2%
81down from #7 since Oct 13, 2026DeepSeek V4.1 Flashas “deepseek-v4.1-flash-max”81.1%
91down from #8 since Oct 13, 2026GPT-5.6 Solas “gpt-5.6-sol-max”81.0%
101down from #9 since Oct 13, 2026GPT-5.5as “gpt-5.5-xhigh”80.2%
111down from #10 since Oct 13, 2026Claude Opus 5as “claude-opus-5-max-effort”80.1%
121down from #11 since Oct 13, 2026Smaug Agenticas “smaug-agentic”79.5%
131down from #12 since Oct 13, 2026GPT-6 Solas “gpt-6-sol-max”79.3%
141down from #13 since Oct 13, 2026Kimi K3as “kimi-k3”79.2%
151down from #14 since Oct 13, 2026Gemini 3.7 Flashas “gemini-3.7-flash-high”78.8%
161down from #15 since Oct 13, 2026Qwen3.8 Max (0902)as “qwen3.8-max”78.5%
171down from #16 since Oct 13, 2026Grok 4.6as “grok-4.6”78.0%
181down from #17 since Oct 13, 2026GPT-5.4as “gpt-5.4-xhigh”78.0%
191down from #18 since Oct 13, 2026Muse Spark 1.2as “muse-spark-1.2-xhigh”78.0%
201down from #19 since Oct 13, 2026GPT-5.6 Terraas “gpt-5.6-terra-max”77.9%
211down from #20 since Oct 13, 2026Claude Sonnet 5.5as “claude-sonnet-5-5-xhigh-effort”77.8%
221down from #21 since Oct 13, 2026DeepSeek V4 Pro 0813as “deepseek-v4-pro-0813”77.4%
221down from #21 since Oct 13, 2026Smaug Flashas “smaug-flash”77.4%
241down from #23 since Oct 13, 2026Grok 4.7as “grok-4.7-xhigh”77.4%
251down from #24 since Oct 13, 2026Gemini 3.1 Pro Previewas “gemini-3.1-pro-preview-high”77.0%
About this board and its history

What it measures

Contamination-limited questions refreshed regularly: math, reasoning, coding, data analysis, language, instruction following.

Rank at each month's end

12345678910NovDecJan 26FebMarAprSepOctNow1Claude Fable 5.12Claude Opus 5.53Claude Fable 54GPT-6 Astra5GPT-6.1 Sol6Muse Spark 1.37210GPT-5.52
Claude Fable 5.1Claude Opus 5.5Claude Fable 5GPT-6 AstraGPT-6.1 SolMuse Spark 1.3Claude Opus 4.5gpt-5-1-codexGPT-5.5GPT-5.4

Other reasoning boards

Source

Scores as published by LiveBench on the capture date (2026-11-09). Source: livebench.ai release tables.

livebench.ai/