Integration guide · Coding assistant

Continue on EU-hosted open models

Add Frontière AI models to Continue's config.yaml with provider openai and apiBase, and use them for chat, edits and agent mode. ~/.continue/config.yaml: a models entry with provider: openai and apiBase; the key in ~/.continue/.env.

Settings from the official documentation, checked on 12 October 2026. Not run end to end by us.

API format

OpenAI-compatible

Base URL

https://getfrontiereai.eu/api/v1

Where it goes

~/.continue/config.yaml: a models entry with provider: openai and apiBase; the key in ~/.continue/.env.

Your prompts

Not stated by the vendor

Setup

# ~/.continue/config.yaml
name: Frontière AI
version: 1.0.0
schema: v1
models:
  - name: GLM 5.2 (Frontière AI)
    provider: openai
    model: glm-5.2
    apiBase: https://getfrontiereai.eu/api/v1
    apiKey: ${{ secrets.FRONTIERE_API_KEY }}
    roles: [chat, edit, apply]
    capabilities: [tool_use]
    defaultCompletionOptions:
      contextLength: 256000
      maxTokens: 8192

# ~/.continue/.env
# FRONTIERE_API_KEY=sk-front-...

Which model to pick

24 live models have tool calling measured by a real call through the gateway. Models labeled EU sovereign run at European providers; fast-access models are served through US-controlled infrastructure. Prices are per 100,000 tokens, margin and VAT included.

Good to know

• Autocomplete needs a fill-in-the-middle model on a completions endpoint; Frontière AI serves chat completions only, so keep autocomplete on another model.
• An embedding model such as qwen3-embedding-8b can be added with roles: [embed] for codebase retrieval.
• Anonymous telemetry goes to PostHog unless you turn off Allow Anonymous Telemetry.

Where your prompts go

Continue's docs do not say in so many words where requests leave from. Its local configuration lives on your machine and its air-gapped guide works without internet once telemetry is off, which points to direct calls; hosted Continue agents are a separate product.

On our side, a model labeled EU sovereign runs at a European provider, and the gateway itself runs in Paris on Vercel, a US company — our privacy page says so. Billing keeps the model, the token counts and the amount, never the prompt or the reply.

Sovereignty →Privacy →

Sources

Official documentation, checked on 12 October 2026:

• Continue — OpenAI provider
• Continue — config.yaml reference
• Continue — Model capabilities

FAQ

Other guides

• Claude CodeCoding agent
• OpenCodeCoding agent
• ClineCoding agent
• Kilo CodeCoding agent