Skip to main content

Model setup

info

This page recommends models and providers for Autocomplete. Read more about how to set up your config.yaml here.

Best overall experience​

For the best Autocomplete experience, we recommend using Codestral through the Mistral API. This model offers high-quality completions with an excellent understanding of code context:

config.yaml
models:
  - name: Codestral
    provider: mistral
    model: codestral-latest 
    apiKey: <YOUR_CODESTRAL_API_KEY>
    roles:
      - autocomplete
Codestral API Key

The API keys for Codestral and the general Mistral APIs are different. If you are using Codestral, you probably want a Codestral API key, but if you are sharing the key as a team or otherwise want to use api.mistral.ai, then make sure to set "apiBase": "https://api.mistral.ai/v1" in your tabAutocompleteModel.

Local, offline / self-hosted experience​

For those preferring local execution or self-hosting,Qwen2.5-Coder 1.5B offers a good balance of performance and quality for most users (e.g. using Ollama or LM Studio).

config.yaml
models:
  - name: Qwen2.5-Coder 1.5B
    model: qwen2.5-coder:1.5b-base 
    provider: ollama
    roles:
      - autocomplete

Have more compute? Use qwen2.5-coder:7b-base for potentially higher-quality suggestions.

note

For LM Studio users, navigate to the "My Models" section, find your desired model, and copy the path (e.g., Qwen/Qwen2.5-Coder-1.5B-Instruct-GGUF/qwen2.5-coder-1.5b-instruct-q4_k_m.gguf). Use this path as the model value in your configuration.

Using a remote instance​

When using a remote instance you need to set the "apiBase" value in the configuration:

config.yaml
models:
  - name: Qwen2.5-Coder 1.5B
    model: qwen2.5-coder:1.5b-base
    provider: ollama
    apiBase: http://<my endpoint>:11434
    roles:
      - autocomplete

Other experiences​

There are many more models and providers you can use with Autocomplete. Check them out here. You can also see more examples of Autocomplete configuration here