Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Editing on Aider whole format
Loading...
1.5
Average Tries
CodeLlama-7B-Instruct
-1.208
17.071
35.35
53.629
Jan 22, 2026
Average Tries
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Tries
CodeLlama-7B-Instruct
Size=7B
2026.01
1.5
CodeLlama-70B-Instruct
Size=70B
2026.01
15
OpenCoder-8B-Instruct
Size=8B
2026.01
30.8
Llama-3.1-8B-Instruct
Size=8B
2026.01
33.1
StarCoder2-15B-Instruct
Size=15B
2026.01
38.2
CodeQwen1.5-7B-Chat
Size=7B
2026.01
38.3
DeepSeek-Coder-6.7B-Instruct
Size=6.7B
2026.01
44.4
Seed-Diffusion-Preview(0705)
Size=-
2026.01
44.4
Codestral-22B
Size=22B
2026.01
51.1
DeepSeek-Coder-V2-Lite-Instruct
Size=2.4B/16B
2026.01
52.6
Yi-Coder-9B-Chat
Size=9B
2026.01
54.1
DeepSeek-Coder-33B-Instruct
Size=33B
2026.01
54.5
Stable-DiffCoder-8B-Instruct
Size=8B
2026.01
54.9
Qwen3-8B
Size=8B
2026.01
55.6
Seed-Coder-8B-Instruct
Size=8B
2026.01
57.1
Qwen2.5-Coder-7B-Instruct
Size=7B
2026.01
57.9
Qwen2.5-Coder-14B-Instruct
Size=14B
2026.01
69.2
Feedback
Search any
task
Search any
task