search

princeton-nlp / SWE-agent

SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It solves 12.47% of bugs in the SWE-bench evaluation set and takes just 1 minute to run.
https://princeton-nlp.github.io/SWE-agent/
MIT License
11.86k stars 1.18k forks source link

How to run SWE-agent on gpt 3.5? #605

Closed ivan4722 closed 1 week ago

ivan4722 commented 1 week ago

Describe the issue

Hi, I modified the command from the documentation to try and run on 3.5, but it does not run. I tried

python run.py --model_name gpt-3.5 --per_instance_cost_limit 2.00 --config_file ./config/default.yaml --split test

and 

python run.py --model_name gpt3.5 --per_instance_cost_limit 2.00 --config_file ./config/default.yaml --split test

and 

python run.py --model_name gpt-3.5-turbo --per_instance_cost_limit 2.00 --config_file ./config/default.yaml --split test

Optional: Relevant documentation page

No response

ofirpress commented 1 week ago

You have the list of available models here: https://github.com/princeton-nlp/SWE-agent/blob/d71f1f5eeb7631acdbdbfaa0938c6519af2cbc70/sweagent/agent/models.py#L179