Continue integration

Add Modellane to the Continue extension for VS Code and JetBrains with a short config entry for chat and edit models.

Continue reads its models from a config file. One entry with the Modellane base URL and key makes our models available in chat and edit.

Continue is an open-source coding assistant for VS Code and JetBrains IDEs. It reads its models from a config file, and one entry makes Modellane models available for chat, edit and apply through the Chat Completions API (POST /v1/chat/completions).

Add the model#

Open your Continue config file (config.yaml, reachable from the settings in the Continue panel) and add an entry under models:

YAML

name: My Config
version: 1.0.0
schema: v1

models:
  - name: Modellane
    provider: openai
    model: lane-1
    apiBase: https://usemodellane.com/v1
    apiKey: <your-api-key>
    useResponsesApi: false
    roles:
      - chat
      - edit
      - apply

Replace <your-api-key> with a key from API keys. If you keep the config in a shared repository, store the key as a Continue secret and reference it with the secrets syntax from the Continue documentation instead of pasting it. To add more models, repeat the entry with another id from Models & pricing.

Context and output size#

Continue estimates how much code it can send from the model's context length. If it cannot detect the size for a custom model, set defaultCompletionOptions with contextLength and maxTokens to the values from the models table.

Verify#

Open the Continue chat, select the Modellane model and send "Reply with the word ready." A reply means the entry works. An authentication error points at the key, and model_not_found at the model id.