-
Notifications
You must be signed in to change notification settings - Fork 37
2.3.13 Satellite: aider
av edited this page Sep 14, 2024
·
1 revision
Handle:
aider
URL: http://localhost:34011/
Aider lets you pair program with LLMs, to edit code in your local git repository. Start a new project or work with an existing git repo. Aider works best with GPT-4o & Claude 3.5 Sonnet and can connect to almost any LLM.
Harbor runs aider
in a CLI mode.
# [Optional] pre-pull if needed
harbor pull aider
# See available options
harbor aider --help
When running aider
, Harbor will mount current working directory as a container workspace. This means that running aider
from a subfolder in a git repo will mask away the fact it's a git repo within the container.
# Run aider in the current directory
harbor aider
# Aider supports browser/GUI mode
# "Network URL" will be open-able in your browser
harbor aider --gui
aider
is pre-configured to automatically work with the LLM backends supported by Harbor. You only need to point it to the correct model.
# Lists active services
harbor ls -a # nothing
# Will try using OpenAI API
harbor aider
# Run one of supported backends
harbor up vllm webui
harbor ls -a # webui vllm
# Set a new model for aider as the model that
# vllm is currently using
harbor aider model $(harbor vllm model)
# Or set a specific model
harbor aider model "microsoft/Phi-3.5-mini-instruct"
# Will use vLLM model
harbor aider
# Let's run with Ollama
harbor down
# "ollama" specifier is not needed if ollama
# is one of the default services
harbor up ollama
# Will use Ollama config now, ensure to point
# it to the correct model!
harbor aider
You can configure aider in Harbor via:
- specifying supported CLI args
-
aider/override.env
file, official .env reference -
aider/confgis/aider.config.yml
file, official config reference- Note, that Harbor will override certain portions of this config to connect Aider to the LLM backends