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

v0.1.63.a15b497a47f2浙ICP备09076226号-7

Back to benchmarks

DeepSWE 1.1Coding

Official website

113 long-horizon software engineering tasks, averaged over five runs.

Ranking Sources

#ModelScore

1

Muse WonderMeta

78.8%

2

Gemini 4 ArgonGoogle

77.9%

3

DeepSeek-V4.1-Flash-1017DeepSeek

77.0%

4

Muse Spark 1.3Meta · max mode

75.4%

5

GPT-6.1 SolOpenAI · high

75.2%

6

Claude Opus 5.5Anthropic

74.2%

7

DeepSeek-V4.1-FlashDeepSeek

74.2%

8

GPT-6 AstraOpenAI · xhigh

74.1%

9

Claude Opus 5Anthropic

74.0%

10

Gemini 3.8 FlashGoogle

73.8%

51 models total

Benchmark details loading

Back to benchmarks

DeepSWE 1.1Coding

Official website

113 long-horizon software engineering tasks, averaged over five runs.

Ranking Sources

#ModelScore

1

Muse WonderMeta

78.8%

2

Gemini 4 ArgonGoogle

77.9%

3

DeepSeek-V4.1-Flash-1017DeepSeek

77.0%

4

Muse Spark 1.3Meta · max mode

75.4%

5

GPT-6.1 SolOpenAI · high

75.2%

6

Claude Opus 5.5Anthropic

74.2%

7

DeepSeek-V4.1-FlashDeepSeek

74.2%

8

GPT-6 AstraOpenAI · xhigh

74.1%

9

Claude Opus 5Anthropic

74.0%

10

Gemini 3.8 FlashGoogle

73.8%

51 models total