ModelsCompareBest forBenchmarksStatusPricingAPI

What is LMArena (Chatbot Arena)?

A human-preference leaderboard: people compare two anonymous models' answers blind and vote, producing an Elo rating from millions of pairwise votes.

LMArena (Chatbot Arena) scores by model

1
Claude Opus 4.6Anthropic1497.5 ↗
2
Claude Opus 4.7Anthropic1483.4 ↗
3
Qwen3.8-MaxQwen1480.6 ↗
4
Gemini 3.1 Pro PreviewGoogle1480.1 ↗
5
GLM-5.3-FlashZ.ai1471.9 ↗
6
ERNIE 5.1Baidu1468.2 ↗
7
GPT-5.5OpenAI1465.6 ↗
8
MiMo-V2.5-ProXiaomi MiMo1464.7 ↗
9
GLM-5.1Z.ai1462.4 ↗
10
Claude Sonnet 4.6Anthropic1458.3 ↗
11
Gemini 2.5 ProGoogle1457.8 ↗
12
Kimi K2.6Moonshot AI1454.9 ↗
13
Qwen3.7-PlusQwen1454.2 ↗
14
Claude Opus 4.8Anthropic1452.7 ↗
15
GPT-5.4OpenAI1452.6 ↗
16
DeepSeek V4 ProDeepSeek1450.6 ↗
17
Grok 4.5xAI1450.1 ↗
18
Dola Seed 2.0 ProByteDance Seed1447.6 ↗
19
GLM-5Z.ai1446.3 ↗
20
NVIDIA Nemotron 3 Ultra 550B-A55B NVFP4NVIDIA1444.7 ↗
21
Gemma 4 31BGoogle1441.7 ↗
22
Hy3Tencent Hunyuan1440.6 ↗
23
GLM-4.6Z.ai1440.5 ↗
24
Qwen3.8 27BQwen1439.3 ↗
25
Qwen3.5 397B-A17BQwen1438.3 ↗
26
Qwen3.6-PlusQwen1436.7 ↗
27
GLM-4.7Z.ai1435.9 ↗
28
Gemini 3.5 Flash-LiteGoogle1435.5 ↗
29
Gemma 4 26B A4BGoogle1434.5 ↗
30
MiniMax M3MiniMax1433.5 ↗
31
GLM-4.5Z.ai1430.2 ↗
32
MiMo-V2.5Xiaomi MiMo1427.4 ↗
33
Mistral Large 3Mistral AI1427 ↗
34
DeepSeek V3.2DeepSeek1424.8 ↗
35
LongCat Flash ChatMeituan LongCat1422.9 ↗
36
GPT-5.1OpenAI1422.6 ↗
37
Mistral Medium 3.5Mistral AI1420.6 ↗
38
Qwen3.5 122B-A10BQwen1417.9 ↗
39
Gemini 2.5 FlashGoogle1417.3 ↗
40
GPT-5.2OpenAI1412.4 ↗