Skip to main content

leaderboard.cn

Explore

Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio

Library

Models
Pricing
Organizations
Benchmarks
Data sources

leaderboard.cn

Benchmark details loading

Loading page content

About Ranking methodology Changelog

v0.1.26.a85cb1563d03浙ICP备09076226号-7

Back to benchmarks

Arena Agent SteerabilityAgents

Results for 31 models collected from 2 public sources.

#ModelScore

1

GPT-5.6 SolOpenAI · xHigh mode

17.3%

2

Claude Fable 5Anthropic · High mode

12.1%

3

Claude Opus 4.8Anthropic · Thinking mode

10.3%

4

Claude Opus 4.7Anthropic · Thinking mode

9.3%

5

GPT-5.5OpenAI · High mode

8.7%

6

GPT-5.4OpenAI · High mode

8.0%

7

Claude Opus 4.6Anthropic

6.7%

8

Claude Sonnet 5Anthropic · High mode

4.9%

9

GLM-5.2Z.ai · Max mode

4.1%

10

Gemini 3.1 Pro PreviewGoogle · Preview mode

1.8%

31 models total