Continue

An open-source coding assistant for VS Code and JetBrains, configured in config.yaml.

Connect

name: Plugboard
version: 1.0.0
schema: v1
models:
  - name: Plugboard fast (chat)
    provider: openai
    model: fast
    apiBase: http://127.0.0.1:4010/v1
    apiKey: <your app key>
    useResponsesApi: false
    capabilities:
      - tool_use
      - image_input
  - name: Plugboard fast (responses)
    provider: openai
    model: fast
    apiBase: http://127.0.0.1:4010/v1
    apiKey: <your app key>
    useResponsesApi: true
    capabilities:
      - tool_use
      - image_input

• The first entry uses Chat Completions, the second the Responses API. Keep the one you need.
• apiBase ends in /v1.
• Continue can't detect capabilities for an alias, so list them: tool_use (needed for Agent mode) and image_input, only if the model really supports them.
• The file holds the key in plain text; keep it out of version control.

The file is in examples/continue/.

What works

Chat with streaming, Agent mode tool calls and images. useLegacyCompletionsEndpoint: true switches an entry to legacy completions.

Limits

• A model named like gpt-5 or an o-series model uses the Responses API by default; set useResponsesApi: false to keep it on Chat Completions.