Skip to content
LLMFLUX

Get started

Base URLs

Use the correct SDK base and avoid a duplicated /v1 prefix.

SDK base URLs

ClientBase URLFinal request
OpenAI SDKhttps://api.llmflux.dev/v1/v1/chat/completions or /v1/responses
Anthropic SDK / Claude Codehttps://api.llmflux.dev/v1/messages

The OpenAI base includes /v1. The Anthropic SDK adds it itself. Direct HTTP requests use the full endpoint URL, regardless of SDK conventions.

Local development

The backend mounts inference at /v1 and defaults to port 4000. Local OpenAI clients use http://localhost:4000/v1; Anthropic clients use http://localhost:4000. Confirm the configured port for your deployment. The docs and dashboard origins are not inference base URLs.

Verify your connection

curl https://api.llmflux.dev/v1

curl https://api.llmflux.dev/v1/models \
  -H "Authorization: Bearer $LLMFLUX_API_KEY"

GET /v1 provides unauthenticated endpoint discovery; GET /v1/models requires your issued key.