Continue via Tokonomix

//What is Continue?

Continue is an open-source AI code assistant for VS Code and JetBrains. It has four surfaces — Chat, Autocomplete, Edit and an Agent — and you configure exactly which model powers each from a single config.yaml.

• Bring your own models: every surface points at a model you define in config.yaml.
• The openai provider with a custom apiBase routes any surface to an OpenAI-compatible endpoint.
• Assign different models to chat, edit, autocomplete and the agent independently.
• Open source and local-config-first — your keys and routing live in your own file.

//Why route Continue through Tokonomix?

Because Continue lets you assign a model per role, it pairs naturally with Tokonomix: one key, every family, and consensus available for the role where being wrong is expensive.

• Point the chat and edit roles at tokonomix-consensus for verified answers, and autocomplete at a fast cheap model.
• One key, 130+ models — switch the model on any role by editing one line.
• Per-call spend caps keep an over-eager agent within budget.
• Sovereignty routing by hosting_region / origin_country for regulated repos.

//Connect Tokonomix in config.yaml

Add a Tokonomix model to config.yaml with the openai provider and a custom apiBase. Reference it from any role (chat, edit, autocomplete, agent).

models:
  - name: Tokonomix Consensus
    provider: openai
    model: tokonomix-consensus
    apiBase: https://tokonomix.ai/api/v1
    apiKey: tok_live_your-key
    roles:
      - chat
      - edit
  - name: Tokonomix Fast
    provider: openai
    model: anthropic/claude-sonnet-4-6
    apiBase: https://tokonomix.ai/api/v1
    apiKey: tok_live_your-key
    roles:
      - autocomplete

//Match a model to each role

//Verify it works

• Save config.yaml and reload Continue (it hot-reloads the config).
• Open Chat, select the Tokonomix model, and ask a question.
• Confirm the call and token usage appear in your Tokonomix dashboard.