Mistral Vibe VS Code extension with local models

Mistral Vibe VS Code extension stores its configuration in ~/.vibe/config.toml and it shares its settings with Mistral Vibe CLI. On Microsoft Windows, Terminal understands ~/.vibe, and if it does not, go to c:\users\<your user>\.vibe.

Note that if you work with a remote, the configuration of the extension is on the remote computer.

Providers configuration

To use the VS Code extension with local models, or any 3rd party for that matter (OpenAI, Anthropic, …), edit ~/.vibe/config.toml and add a providers section like so:

[[providers]]
name = "llama.cpp"
api_base = "http://localhost:8080/v1"
api_key_env_var = "LLAMACPP_API_KEY"
api_style = "openai"
backend = "generic"
reasoning_field_name = "reasoning_content"
project_id = ""
region = ""

[providers.extra_headers]

If you do not need an API key, leave the field api_key_env_var empty.

If you need an API key, then edit ~/.vibe/.env like so:

LLAMACPP_API_KEY='myapikey'

Check the documentation for all the settings.

Models configuration

Finally, define one or more models for this provider like so:

[[models]]
name = "qwen"
provider = "llama.cpp"
alias = "qwen"
temperature = 1.0
input_price = 1.5
output_price = 7.5
thinking = "high"
auto_compact_threshold = 200000

Restart VS Code to load the new environment variable, and you should now be able to use your new model(s). It will appear in the Model list as the alias.

Troubleshooting

If you cannot select the model, check the api_key_env_var matches the name defined in .env.