UnifyBench

RankingLMArena Text Arena

LMArena Text Arena

Data updated 22 Oct 2026

Bucket
Human pref
Unit
elo
Direction
Higher is better
Version
—
Display harness
LMArena Text
Board
https://lmarena.ai/leaderboard

Compare published benchmark results with category weights →

Models

Results are ordered by score in the display harness, followed by models with no result. Other harnesses are listed separately below. Missing results remain unknown.

26–50 of 471 entries

ModelScoreEvidenceSource-recorded date
Grok 4.6xAI1455official board2026-10-20
Gemini 3.5 Flash-LiteGoogle1455official board2026-10-20
Claude Sonnet 4.5Anthropic1455official board2026-10-15
Hy3Tencent1453official board2026-10-20
MiMo-V2.6-FlashXiaomi1449official board2026-10-20
Step 5 PreviewStepFun1446official board2026-10-20
Gemini 2.5 ProGoogle1445official board2026-10-20
Qwen3.6-plusQwen1443official board2026-10-20
Grok-4.3xAI1442official board2026-10-15
Qwen3.5-397B-A17BQwen1441official board2026-10-20
GLM-4.7Z.ai1441official board2026-10-20
MiniMax-M3MiniMax1441official board2026-10-20
GPT-5.1OpenAI1439official board2026-10-20
Qwen3.8-27BQwen1437official board2026-10-20
DeepSeek-V4-Flash-0731DeepSeek1436official board2026-10-20
GPT-5.2OpenAI1435official board2026-10-20
MiMo-V2.5Xiaomi1434official board2026-10-20
GLM-5V-TurboZ.ai1433official board2026-10-20
Mistral Large 4Mistral1433official board2026-10-20
Mistral Medium 3.5Mistral1426official board2026-10-20
NVIDIA Nemotron 3 UltraNVIDIA1426official board2026-10-15
DeepSeek-V3.2DeepSeek1425official board2026-10-20
GLM-4.6Z.ai1424official board2026-10-20
Qwen3-235B-A22B-Instruct-2507Qwen1422official board2026-10-15
DeepSeek-R1-0528DeepSeek1421official board2026-10-15