Modelflare

Stable, lower-cost AI API gateway for LLM, image, and video models. One key for OpenAI-compatible Chat & Responses, Anthropic Messages, plus native image and video contracts — with ordered model-group fallbacks, per-request usage & cost tracking, Smart keys, and cache hit guarantee.

What you get

• One key for text, image, and video models
• OpenAI-compatible chat — point your SDK at https://modelflare.dev/v1
• Group failover so one unavailable route does not fail the request
• Per-request token, image, and video cost records
• Smart keys and cache hit guarantee
• Guides for Codex, Claude Code, Gemini CLI: see the Coding Agents Space and the Blog

Quick start

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_MODELFLARE_API_KEY",
    base_url="https://modelflare.dev/v1",
)
r = client.chat.completions.create(
    model="gpt-5.6-sol",  # must be enabled on your group
    messages=[{"role":"user","content":"Hello from Hugging Face"}],
)
print(r.choices[0].message.content)

Create a key in the console, then switch base_url to https://modelflare.dev/v1. Full guides: docs.modelflare.dev · SDK walkthrough: openai-sdk · recipes: api-recipes · agents: coding-agents · blog: Modelflare/blog.

This Hub organization hosts Modelflare demos, examples, and community Spaces. Feedback welcome in the start Space discussion #1. Interactive Gradio Chat Demo ships after HF PRO; until then use the console + API path above.