Tools & IDEs

Configure Continue

Configure Continue for VS Code/JetBrains with cheapkeyai.com, config.yaml, environment variables, and autocomplete.

Continue reads configuration from the `.continue` folder. Declare multiple cheapkeyai.com models for chat, edit, apply, and autocomplete.

Endpoint

Base URL 
https://cheapkeyai.com/v1

Configuration Directory

Paths 
macOS / Linux: ~/.continue
Windows: %USERPROFILE%\.continue

Minimal config.yaml

config.yaml 
name: cheapkeyai.com
version: 1.0.0
schema: v1

models:
  - name: GPT-5.2
    provider: openai
    model: gpt-5.2
    apiKey: sk-your-api-key
    apiBase: https://cheapkeyai.com/v1
    roles:
      - chat
      - edit
      - apply
    capabilities:
      - tool_use
      - image_input

  - name: Claude Sonnet 4.5
    provider: anthropic
    model: claude-sonnet-4-5
    apiKey: sk-your-api-key
    apiBase: https://cheapkeyai.com/v1
    roles:
      - chat
      - edit
      - apply

Using Environment Variables

Shell 
export CHEAPKEYAI_API_KEY="sk-your-api-key"
config.yaml with env 
models:
  - name: GPT-5.3 Codex
    provider: openai
    model: gpt-5.3-codex
    apiKey: ${{ env.CHEAPKEYAI_API_KEY }}
    apiBase: https://cheapkeyai.com/v1
    roles: [chat, edit, apply]
    capabilities: [tool_use]

Autocomplete

Autocomplete should use low-latency models to avoid slowing down the editor. If Continue supports a separate tab for autocomplete, declare a lighter model for the autocomplete role.

Troubleshooting

  • Continue does not detect new config: reload window or restart IDE.
  • Authentication error: check if environment variables exist in the IDE process.
  • Model not responding: try cURL directly to `/v1/chat/completions` to separate IDE errors from API errors.