Every AI leaderboard

28 leaderboards · 986 models ranked

LMArena Text Arena leaderboard: who is #1

LMArena · Arena rating, higher is better · 84 models · published Nov 6, 2026

LMArena

Top 25 of 84 models, best variant per model as published

#ModelScoreCorrect
1New, entered at #1 since Oct 29, 2026Gemini 4 Carbonas “Gemini 4 Carbon (High)”1,526
21down from #1 since Oct 29, 2026Gemini 4 Argonas “Gemini 4 Argon (High)”1,523
31up from #4 since Nov 7, 2026Muse Wonderas “Muse Wonder (Max)”1,513
41down from #3 since Nov 7, 2026Claude Opus 5.5as “Claude Opus 5.5 (High)”1,505
51up from #6 since Nov 7, 2026Claude Fable 5as “Claude Fable 5 (High)”1,504
61down from #5 since Nov 7, 2026Claude Fable 5.1as “Claude Fable 5.1 (Max)”1,504
71down from #6 since Oct 29, 2026Claude Opus 4.6as “Claude Opus 4.6 (High)”1,503
81down from #7 since Oct 29, 2026Claude Opus 4.7as “Claude Opus 4.7 (High)”1,501
91down from #8 since Oct 29, 2026Gemini 3.8 Flashas “Gemini 3.8 Flash (High)”1,498
101down from #9 since Oct 29, 2026Muse Spark 1.3as “Muse Spark 1.3 (Max)”1,495
111down from #10 since Oct 29, 2026Muse Spark 1.2as “Muse Spark 1.2 (xHigh)”1,493
121down from #11 since Oct 29, 2026Claude Opus 5as “Claude Opus 5 (High)”1,490
131down from #12 since Oct 29, 2026Muse Spark 1.1Meta1,490
141up from #15 since Nov 7, 2026Muse SparkMeta1,489
151down from #14 since Nov 7, 2026Gemini 3.7 Flashas “Gemini 3.7 Flash (High)”1,489
161down from #15 since Oct 29, 2026Kimi K3as “Kimi K3 (Max)”1,488
171down from #16 since Oct 29, 2026Gemini 3.1 Pro PreviewGoogle1,487
181down from #17 since Oct 29, 2026Gemini 3 ProGoogle1,486
191down from #18 since Oct 29, 2026GPT-5.6 Solas “GPT-5.6 Sol (xHigh)”1,485
201up from #21 since Nov 7, 2026Qwen3.8 Max (0902)as “Qwen3.8-Max”1,483
213up from #24 since Nov 7, 2026GPT-6.1 Solas “GPT-6.1 Sol (Max)”1,483
222down from #20 since Nov 7, 2026Gemini 3.6 Flashas “Gemini 3.6 Flash (High)”1,483
232up from #25 since Nov 7, 2026GLM-5.3as “GLM-5.3 (Max)”1,482
242down from #22 since Nov 7, 2026Claude Opus 4.8as “Claude Opus 4.8 (High)”1,482
252down from #23 since Nov 7, 2026GPT-5.5as “GPT-5.5 (High)”1,481
About this board and its history

What it measures

Blind pairwise votes from millions of people chatting with two anonymous models. Style control on.

Rank at each month's end

12345678910NovDecJan 26FebMarAprMayJunJulAugSepOctNow1Gemini 4 Carbon2Gemini 4 Argon3Muse Wonder4Claude Opus 5.55Claude Fable 56Claude Fable 5.19787Claude Opus 4.6

Source

Scores as published by LMArena on the capture date (2026-11-09). Source: LMArena leaderboard dataset on Hugging Face.

arena.ai/leaderboard/text