Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Automated Program Repair on QuixBugs-Java 40 bugs
Loading...
100
Pass@1 Rate
ChatRepair
61
71.125
81.25
91.375
May 9, 2026
Pass@1 Rate
Updated 22d ago
Evaluation Results
Method
Method
Links
Pass@1 Rate
ChatRepair
Backbone=GPT-3.5-turbo
2026.05
100
+ Stage III (PPO, Rseq only)
Params=32B
2026.05
95
BOOSTAPR (+ Rline)
Params=32B
2026.05
95
Lingma SWE-GPT
Backbone=Qwen2.5-72B,...
2026.05
92.5
SWE-Fixer
Backbone=Qwen2.5-72B,...
2026.05
92.5
+ Stage I (SFT)
Params=32B
2026.05
92.5
SWE-Gym
Backbone=Qwen2.5-Coder...
2026.05
90
SWE-RL
Backbone=Llama-3-70B,...
2026.05
90
Qwen2.5-Coder-32B (base)
Params=32B
2026.05
90
Agentless
Backbone=GPT-4o
2026.05
87.5
SWE-agent
Backbone=Claude 3.5 So...
2026.05
85
AutoCodeRover
Backbone=GPT-4o
2026.05
82.5
RLEF
Backbone=Llama-3-8B, P...
2026.05
80
RepairLLaMA
Backbone=CodeLlama-7B,...
2026.05
75
CodeRL
Backbone=CodeT5-large,...
2026.05
67.5
KNOD
Backbone=CodeT5-base,...
2026.05
62.5
Feedback
Search any
task
Search any
task