Model Horizon
DashboardModelsCompareBenchmarks
© 2026 Model Horizon
About|Terms
SYS.v0.1.0
Skip to content
  1. Home
  2. /Benchmarks
  3. /Arena Elo

Arena Elo Leaderboard

Chatbot Arena Elo Rating

Crowdsourced ranking where users compare model outputs head-to-head. Higher Elo indicates stronger overall performance across diverse tasks.

26Models Tested
1,500Highest Score
1,412.2Average
310Spread
#ModelProviderScore
1Gemini 3.1 ProGoogle
GGoogle
1,500
Try
2Claude Opus 4.6Anthropic
AAnthropic
1,496
Try
3Gemini 3 ProGoogle
GGoogle
1,486
Try
4GPT-5.3 CodexOpenAI
OOpenAI
1,480
Try
5Claude Sonnet 4.6Anthropic
AAnthropic
1,470
Try
6Gemini 3 FlashGoogle
GGoogle
1,470
Try
7Claude Opus 4.5Anthropic
AAnthropic
1,467
Try
8Grok 4.1xAI
XxAI
1,465
Try
9GPT-5.1OpenAI
OOpenAI
1,458
Try
10Claude Sonnet 4.5Anthropic
AAnthropic
1,450
Try
11Gemini 2.5 ProGoogle
GGoogle
1,450
Try
12GPT-5.2OpenAI
OOpenAI
1,438
Try
13o3OpenAI
OOpenAI
1,433
Try
14DeepSeek V3.2DeepSeek
DDeepSeek
1,419
Try
15DeepSeek R1DeepSeek
DDeepSeek
1,418
Try
16Mistral Large 3Mistral
MMistral
1,414
Try
17GPT-4.1OpenAI
OOpenAI
1,413
Try
18Gemini 2.5 FlashGoogle
GGoogle
1,410
Try
19Grok 4xAI
XxAI
1,410
Try
20Claude Haiku 4.5Anthropic
AAnthropic
1,404
Try
21o4-miniOpenAI
OOpenAI
1,380
Try
22Llama 4 MaverickMeta
MMeta
1,365
Try
23Llama 4 ScoutMeta
MMeta
1,330
Try
24GPT-4.1 miniOpenAI
OOpenAI
1,280
Try
25Mistral Small 3.2Mistral
MMistral
1,220
Try
26GPT-4.1 nanoOpenAI
OOpenAI
1,190
Try