Skip to content
← Back to guides

Continue setup (VS Code / JetBrains)

Continue is an open-source AI coding assistant for VS Code and JetBrains that defines its models in config.yaml. Add a model with provider anthropic and point apiBase at 6090ti to call Claude models with a 6090ti API key.

Before you start

  • Create an account and top up.
  • Create an API key in the console (top up and create a key).
  • Request https://api.6090ti.com/v1/models with that key and pick a model name from the list; below it is written as <model name>. Prices per model: Claude model APIs.

Step 1: Install Continue

Search for Continue in the VS Code Extensions view or the JetBrains plugin marketplace and install it.

Step 2: Edit config.yaml

The local configuration file is ~/.continue/config.yaml (Windows: %USERPROFILE%\.continue\config.yaml); you can also open it from the Continue panel settings. Add under models:

models:
  - name: 6090ti Claude
    provider: anthropic
    model: <model name>
    apiBase: https://api.6090ti.com/v1/
    apiKey: <your key>
    roles:
      - chat
      - edit
      - apply

apiBase ends in /v1/: Continue appends messages, so requests go to https://api.6090ti.com/v1/messages. If the test returns 404, change it to https://api.6090ti.com and try again.

Step 3: Test

Save the file, select 6090ti Claude at the top of the Continue panel and send “Reply with OK only”. If the call appears in the console usage records, the setup works.

Notes

  • The key in config.yaml is plain text. Do not share the file or commit it to a repository.
  • Autocomplete makes very frequent calls; large Claude models are not a good fit for it.

Common errors

What you seeCause and fix
401, INVALID_API_KEYWrong key, or extra spaces. Copy the full key from the console again.
403, INSUFFICIENT_BALANCENot enough balance. Top up and retry.
404 or “model not available”The model name is wrong or not enabled for your key. Check with /v1/models.
404 on the pathThe base URL has a path too many or too few. Enter it exactly as in this guide; do not add /v1/messages yourself.
429Too many requests or the upstream is busy. Wait and retry.

More error details: API reference.