Skip to content
Updates & support on Telegram
UncensoredAI

~/.continue/config.yaml

Native [OI] support

Continue with a custom [OI] endpoint

Continue's config.yaml takes an openai provider with an apiBase override. Point chat, edit and autocomplete at UncensoredAI and pick cheap fast models for inline work.

Base URL

https://uncensoredai.dev/v1

  1. 01

    Edit config.yaml

    Add the provider entries. apiBase must include /v1. Use a cheap fast model for autocomplete and a stronger one for chat.

    ~/.continue/config.yaml
    name: UncensoredAI
version: 1.0.0
schema: v1
models:
  - name: GPT-6 Sol
    provider: openai
    model: gpt-6-sol
    apiBase: https://uncensoredai.dev/v1
    apiKey: sk-ua-XXXX-XXXX-XXXX-XXXX
    roles: [chat, edit, apply]
  - name: DeepSeek V4.1 Flash
    provider: openai
    model: deepseek-v4.1-flash
    apiBase: https://uncensoredai.dev/v1
    apiKey: sk-ua-XXXX-XXXX-XXXX-XXXX
    roles: [autocomplete]
  2. 02

    Reload Continue

    Save the file and Continue reloads automatically. If autocomplete feels slow, keep it on the cheap tier.

FAQ

Can Continue use a non-[OI] API?+

Yes — any [OI]-compatible endpoint via provider: openai with an apiBase override.

Which model is best for Continue autocomplete?+

A fast low-cost tier such as DeepSeek V4.1 Flash. Reserve the reasoning tiers for chat and edits.

More integrations

Generate access keyAll integrations