cv.cm

Best AI models: LLM leaderboard, November 9, 2026

The top 30 chat models on the Arena text leaderboard, ranked by blind human votes. Search, sort by any column or download the CSV.

Key facts

  • First: Gemini 4 Carbon (High) (1,538), then Gemini 4 Argon (High) (1,530) and Claude Opus 5.5 (High) (1,513).
  • Anthropic holds the most places in the top 10: 6.
  • Data as of November 9, 2026.
Download CSV

30 rows

Best AI models: LLM leaderboard, November 9, 2026. Select a column heading to sort.
1Gemini 4 Carbon (High)GoogleProprietary1,5384,282
2Gemini 4 Argon (High)GoogleProprietary1,53019,787
3Claude Opus 5.5 (High)AnthropicProprietary1,51315,249
4Claude Fable 5.1 (Max)AnthropicProprietary1,51220,087
5Claude Opus 5 (Max)AnthropicProprietary1,50635,710
6Muse Wonder (Max)MetaProprietary1,50312,032
7Claude Opus 5 (High)AnthropicProprietary1,50273,697
8Claude Opus 4.6 (High)AnthropicProprietary1,50284,424
9Gemini 3.8 Flash (High)GoogleProprietary1,50061,143
10Claude Opus 4.6AnthropicProprietary1,49789,228
11MiMo-V2.6-ProXiaomiMIT1,4924,066
12Claude Fable 5 (High)AnthropicProprietary1,49245,524
13Muse Spark 1.3 (Max)MetaProprietary1,49126,111
14Claude Opus 4.7 (High)AnthropicProprietary1,49064,308
15Gemini 3.7 Flash (High)GoogleProprietary1,48933,300
16Qwen3.8-MaxAlibabaProprietary1,48536,324
17Muse Spark 1.2 (xHigh)MetaProprietary1,48415,004
18Claude Opus 4.7AnthropicProprietary1,48365,369
19Gemini 3.5 Flash (High)GoogleProprietary1,48347,642
20Gemini 3.1 Pro PreviewGoogleProprietary1,480132,296
21Gemini 3 ProGoogleProprietary1,48041,829
22Gemini 3.6 Flash (High)GoogleProprietary1,47935,405
23Muse Spark 1.1MetaProprietary1,47945,562
24Gemini 3.5 Flash (Medium)GoogleProprietary1,47846,377
25GLM-5.3 (Max)Z.aiMIT1,47631,017
26Qwen3.7-Max-PreviewAlibabaProprietary1,4753,904
27Kimi K3 (Max)Moonshot AIKimi K3 license1,47440,867
28Muse SparkMetaProprietary1,47414,009
29GPT-5.5 (High)OpenAIProprietary1,47168,359
30Qwen3.5-Max-PreviewAlibabaProprietary1,47122,419

Source: Arena leaderboard dataset · Licence: CC BY 4.0 · Data as of November 9, 2026 · Collected 2026-11-10

About this ranking

Arena lets people put a prompt to two anonymous models and vote for the better answer. Those votes become a rating, so a gap of a few points means little: check the votes column, because new models with few votes move most. This table shows the overall category from the latest published snapshot.

A high rating shows what voters preferred in general chat. It does not measure price, speed, or how a model does on your own work, so try the top few on a real task before choosing.

FAQ

What is the best AI model right now?

By Arena's votes as of November 9, 2026, Gemini 4 Carbon (High) ranks first with a score of 1,538, followed by Gemini 4 Argon (High) and Claude Opus 5.5 (High). The order changes often, and models close in score can swap places as votes come in.

How is this LLM leaderboard scored?

People compare two anonymous models on the same prompt and vote. Arena turns the votes into a rating similar to chess ratings. A higher score means a model tends to beat others in those matchups.

Why does one model appear several times?

Arena ranks settings separately, such as a High or Max thinking effort, because the same model can answer differently at each.

More rankings

All rankings