Developers

Build with LLMFlux: an AI API for developers

One key, standard API formats and clear records of every request. This page collects what you need to integrate: endpoints, SDK setup, the CLI, and the guides for errors, limits and streaming.

Start in three steps

  1. Create an account at dashboard.llmflux.dev and add credit or claim the sign-up credit.
  2. Create an API key under API Keys and export it as LLMFLUX_API_KEY.
  3. Send a request to the OpenAI-compatible or Anthropic-compatible endpoint:
curlbash
curl https://api.llmflux.dev/v1/chat/completions \
  -H "Authorization: Bearer $LLMFLUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "messages": [{ "role": "user", "content": "Explain what an API gateway does in one sentence." }]
  }'

Base URLs

FormatBase URLMain endpoints
OpenAI SDKshttps://api.llmflux.dev/v1/chat/completions, /responses, /models
Anthropic SDKs, Claude Codehttps://api.llmflux.dev/v1/messages

Authenticate with Authorization: Bearer <key> or x-api-key: <key>.

Tools for developers

  • Playground

    Chat with any catalogue model in the dashboard to compare answers before writing code.

  • Request log

    Search every request by model and status, with token counts and cost.

  • Usage charts

    Daily usage and spend across all your keys.

  • LLMFlux CLI

    A terminal client with browser sign-in, installed with one command (macOS, Linux, WSL; needs Node.js 18+).

  • Interactive reference

    Try endpoints from the docs with your own key.

  • Launchpad credits

    Students, educators, researchers and startups can apply for credit grants.

Install the CLIbash
curl -fsSL https://llmflux.dev/install.sh | bash

Guides by language and tool

Language guides cover Python, JavaScript and curl. For coding agents and editors, see integrations. To choose models, start from the model catalogue and model routing.

Documentation

Frequently asked questions

How do I get an API key?
Create an account in the dashboard, then open API Keys and create a key. Keys are shown once; store them in an environment variable such as LLMFLUX_API_KEY.
Is there a free way to try it?
New accounts can receive a sign-up credit after verifying with Telegram, which is enough to send real test requests. The playground in the dashboard lets you try models without writing code.
How do I see what a request cost?
Every request is listed in the dashboard's request log with model, tokens, status and cost. GET /v1/usage returns usage and balance from the API.
What are the rate limits?
Limits are applied per key and account and are documented in the rate-limits guide. A 429 response includes Retry-After.

Start with one key

Create an account, generate an API key, and send your first request in a few minutes.