Skip to main content
Evaluation sources/Arena Text Overall
ParticipatingHuman preference & composite

Arena Text Overall

LMArena · Public model evaluation results from Arena Text Overall.

Official evaluation
Source description

This source measures concrete delivery performance in Human preference & composite; raw scores are shown separately from the normalized capability score.

Limitation: Different versions, reasoning settings, and test sets may change scores; read them with the data date.

How it is used

Participating sources affect capability scores; reference, cross-reference, and observing sources are for comparison and verification only.

Model results from this source

Gemini 4 Carbon1538.276819275.52026-11-09
Gemini 4 Argon1528.5769991373.72026-10-28
Claude Opus 5.51515.7998278271.32026-10-28
Claude Fable 5.11513.8339203771.02026-10-28
Claude Opus 51507.9090505769.92026-10-28
Claude Opus 4.61501.7236438468.82026-10-28
Muse Wonder1500.9554786668.62026-10-15
Muse Wonder1500.9554786668.62026-10-15
Muse Wonder1499.1665529168.32026-10-28
Gemini 3.8 Flash1498.4923216468.22026-10-28
Claude Fable 51491.561124566.92026-10-28
MiMo V2.6 Pro1491.3511193466.92026-10-28
Muse Spark 1.31490.6255540366.72026-10-28
Claude Opus 4.71489.719561166.62026-10-28
Qwen3.8 Max1483.2905770765.42026-10-28
Muse Spark 1.21482.6647890565.32026-10-28
Gemini 3.1 Pro1480.1061552964.82026-10-28
Gemini 3.6 Flash1479.7907457564.72026-10-28
Muse Spark 1.11478.2815890864.52026-10-28
Kimi K31475.6886282164.02026-10-28
GLM 5.31473.77570563.62026-10-28
GLM-5.3 Flash1473.0371705163.52026-10-28
Claude Sonnet 5.51472.8510730863.52026-10-28
GPT-5.51471.5686765663.22026-10-28

Page 1 of 7

Frequently asked questions