Continue

Supported setup

Continue can use OpenModel for models that support the Responses protocol. The configuration must explicitly enable Continue's Responses API path; OpenModel does not provide Chat Completions.

Configure Continue

Open %USERPROFILE%\.continue\config.yaml on Windows or ~/.continue/config.yaml on macOS/Linux and add a model:

name: OpenModel
version: 0.0.1
schema: v1

models:
  - name: OpenModel GPT-5.4
    provider: openai
    model: gpt-5.4
    apiBase: https://api.openmodel.ai/v1
    apiKey: om-your-api-key
    useResponsesApi: true

provider: openai and useResponsesApi: true are Continue configuration values. Together they make Continue call POST /v1/responses; they do not mean that OpenModel implements other OpenAI endpoints.

Before changing the model, confirm that its current supported_protocols includes responses on Available Models. Do not use this configuration for a Messages-only model.

Verify

Save the file, reload Continue, select the configured model, and send a short message. If the request reaches /v1/chat/completions, the Responses setting was not applied by the installed Continue version.

Limitations

• This guide covers Continue chat through Responses only.
• Tab autocomplete may use a different endpoint and is not covered by this configuration.
• Model availability and protocol support come from OpenModel's live model catalog, not from this example.

Next steps

• Available Models
• Responses API
• Authentication