New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix(backend) changes to improve Command-R+ behavior, plus file i/o error improvements, attempt 2 #1417
fix(backend) changes to improve Command-R+ behavior, plus file i/o error improvements, attempt 2 #1417
Conversation
…ous file IO errors, added timeout and max return token configurations for the LLM api.
…sandbox when resolving paths in fileio operations, add customizable timeout for bash commands, mention said timeout in llm prompt.
…ng to delete it from the response afterwards, fixed get_working_directory for ssh_box.
…r-plus-changes # Conflicts: # opendevin/action/fileop.py
… command-r-plus-changes # Conflicts: # agenthub/monologue_agent/utils/prompts.py # opendevin/llm/llm.py # opendevin/schema/config.py # tests/test_fileops.py
… command-r-plus-changes # Conflicts: # opendevin/config.py
This integration test failure does not occur locally for some reason, investigating. |
…n local sandbox mode, combine exception handlers in prompts.py.
@rbren I have fixed the integration test issue. It was introduced when the local-sandbox integration test was added late last night 7d5856e . The test environment configuration did not account for how the path inside the sandbox could no longer be /workspace, and had to be adjusted to match the external host path instead. |
I'm sorry about that @computer-whisperer , I looked at the issue and missed that conflict. I think we could get this one in, it has some pretty awesome changes and they seem to help even poor old GPT-3.5! |
#1409