A Github Copilot commits generator free alternative, that run on your device using ollama.
- No telemetry or tracking.
- No API key needed.
- Different models available.
- No Internet connection needed.
- Install Ollama on your local machine.
- Install the model to use:
ollama pull [model_name]
, recommended to usemistral
orcodellama
. In the tests I did, it works better withmistral
. - Make sure ollama is running, you can do it by visiting http://127.0.0.1:11434/ in your web browser (The port number might be different for you). If not, only opening the app should be enough, or run in your terminal:
ollama serve
.
-
Model: You can select the model from the plugin configuration.
mistral
- defaultcodellama
custom
- It allow you to write down the model name that you have set on ollama. -
Use Emojis: It allow you to enable or disable the use of emojis in commit messages.
-
Custom Emojis: It allow you to write down the emojis you want to use in the next template object in the VSCode config.json.
"commitollama.commitEmojis": { "feat": "โจ", "fix": "๐", "docs": "๐", "style": "๐", "refactor": "โป๏ธ", "test": "๐งช", "chore": "๐ฆ", "revert": "โช" }
-
Custom Endpoint: Ollama usually uses port 11434. It is the value that will be used if empty.
-
Custom Summary Prompt: The prompt that will be used to generate the summary of all git diff.
-
Custom Commit Prompt: The prompt that will be used to generate the commit message.
-
Custom Summary Temperature: The temperature that will be used to generate the summary of all git diff.
-
Custom Commit Temperature: The temperature that will be used to generate the commit message.
Sometimes can generate quite long commits, but it give you an idea of what the commit should be and can be edited manually to achieve the correct length.
Added support for emojis.
Added support for custom prompts and llm temperature.
Initial release of commitollama.