API
Skills/Coding

Coding

Code generation, debugging, and software engineering tasks β€” from writing functions to fixing real-world GitHub issues.

109
Models Ranked
75.3
Top Score
32.9
Average Score
7
Benchmarks
1Google DeepMindGoogle DeepMind75.3
2DeepSeekDeepSeek74.2
3OpenAIOpenAI71.6
4AnthropicAnthropic66.1
5OpenAIOpenAI60.4
6AnthropicAnthropic60.4
7OpenAIOpenAI59.1
8OpenAIOpenAI59.1
9OpenAIOpenAI57.4
10AnthropicAnthropic56.9
11DeepSeekDeepSeek56.8
12DeepSeekDeepSeek56.5
13GoogleGoogle56.1
14OpenAIOpenAI55.0
15OpenAIOpenAI54.0
16OpenAIOpenAI54.0
17OpenAIOpenAI53.8
18xAIxAI50.8
19AlibabaAlibaba49.2
20Alibaba QwenAlibaba Qwen48.4
21DeepSeekDeepSeek48.4
22
ZA
z-ai
48.2
23
ZA
z-ai
48.2
24OpenAIOpenAI48.1
25
M
moonshotai
47.9
26
ZA
z-ai
47.4
27
M
moonshotai
47.3
28OpenAIOpenAI47.0
29OpenAIOpenAI47.0
30OpenAIOpenAI47.0
31DeepSeekDeepSeek46.7
32AnthropicAnthropic46.2
33xAIxAI45.9
34xAIxAI45.9
35
M
moonshotai
45.6
36AnthropicAnthropic45.6
37xAIxAI45.3
38xAIxAI45.3
39Google DeepMindGoogle DeepMind45.2
40Google DeepMindGoogle DeepMind43.4
41OpenAIOpenAI43.3
42xAIxAI42.9
43AnthropicAnthropic42.4
44Alibaba QwenAlibaba Qwen41.0
45AnthropicAnthropic39.6
46AnthropicAnthropic39.1
47AnthropicAnthropic39.1
48Google DeepMindGoogle DeepMind38.9
49DeepSeekDeepSeek38.4
50Google DeepMindGoogle DeepMind38.2