Skip to content
Fast Inference
DOCS / Fx
Integration

Fx

Track availability of the planned Fx gateway adapter and use Fast Inference's API directly in the meantime.

Fx is part of the fast-web integration catalog, but a supported public configuration contract and released fast fx adapter are not available yet. This page records the availability boundary so the product does not present an unverified command as working behavior.

Use the API directly

You can still use the same models and project key from any OpenAI-compatible client:

curl
curl https://api.inference.net/v1/chat/completions \
  -H "Authorization: Bearer $INFERENCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [{ "role": "user", "content": "Hello" }]
  }'

The request is billed and displayed in fast-web exactly like traffic from a coding-agent adapter.

What the adapter will require

The integration cannot be documented as released until all of these are true:

  • Fx exposes a stable custom gateway and credential contract;
  • the adapter changes only owned fields and can restore them safely;
  • the selected model is validated against the current catalog;
  • status distinguishes active, modified, and missing values; and
  • an end-to-end request appears in the correct fast-web project.

Follow availability

The Integrations dashboard reflects the current state. Use Codex, OpenCode, or Pi when you need a supported OpenAI-compatible coding-agent route today.