πŸ† Model Leaderboard

Compare performance across dimensions to quickly find the best open model for you

UPDATE
πŸ₯‡
Claude Opus 4.5 New #1
Anthropic
Overall score 73.6 β˜…β˜…β˜…β˜…β˜†
πŸ₯ˆ
Gemini 3 Pro
Google
Overall score 69.8 β˜…β˜…β˜…β˜†β˜†
πŸ₯‰
GPT 5.1 Codex Max
OpenAI
Overall score 69.4 β˜…β˜…β˜…β˜†β˜†
4
Claude Sonnet 4.5
Anthropic
Overall score 69.3 β˜…β˜…β˜…β˜†β˜†
5
Claude 4.1 Opus
Anthropic
Overall score 68.0 β˜…β˜…β˜…β˜†β˜†
6
GPT 5.2
OpenAI
Overall score 68.0 β˜…β˜…β˜…β˜†β˜†
7
Deepseek V3.2
DeepSeek
Overall score 65.3 β˜…β˜…β˜…β˜†β˜†
8
GPT 5.1
OpenAI
Overall score 65.3 β˜…β˜…β˜…β˜†β˜†
9
GPT 5 Mini
OpenAI
Overall score 64.1 β˜…β˜…β˜…β˜†β˜†
10
GPT 5 Pro
OpenAI
Overall score 63.6 β˜…β˜…β˜…β˜†β˜†
11
GPT 5.1 Codex
OpenAI
Overall score 63.4 β˜…β˜…β˜…β˜†β˜†
12
Gemini 3 Flash Minimal
Google
Overall score 63.0 β˜…β˜…β˜…β˜†β˜†
13
Claude 4 Sonnet
Anthropic
Overall score 62.9 β˜…β˜…β˜…β˜†β˜†
14
Claude Haiku 4.5
Anthropic
Overall score 62.7 β˜…β˜…β˜…β˜†β˜†
15
Devstral 2512
Community
Overall score 60.9 β˜…β˜…β˜…β˜†β˜†
16
Gemini 3 Flash
Google
Overall score 58.2 β˜…β˜…β˜…β˜†β˜†
17
Deepseek V3.2 Exp
DeepSeek
Overall score 57.8 β˜…β˜…β˜…β˜†β˜†
18
Kimi K2 Thinking
Moonshot
Overall score 57.5 β˜…β˜…β˜…β˜†β˜†
19
GPT 5.1 Codex Mini
OpenAI
Overall score 57.5 β˜…β˜…β˜…β˜†β˜†
20
Kimi K2
Moonshot
Overall score 57.1 β˜…β˜…β˜…β˜†β˜†
21
GLM 4.6
ζ™Ίθ°± AI
Overall score 56.8 β˜…β˜…β˜…β˜†β˜†
22
Gemini 2.5 Pro 06 05 Thinking
Google
Overall score 56.6 β˜…β˜…β˜…β˜†β˜†

CURRENT MODEL

Claude Opus 4.5

Rank #1

Ranked #1 in the LiveBench Coding evaluation, overall score 73.6.

Arena score: 73.6 (raw 74)
Context window: β€”
Organization: Anthropic
License: β€”
Official price No official price

SELECT ACTIVE GATEWAYS

● Encrypted link verified
All data comes from public testing; actual results may vary by usage and scenario.