One Unified API
A single OpenAI-compatible endpoint. Swap models by changing one string — no new SDKs.
100% open source · no subscriptions
One Open API for
Every Language Model
Route across Groq, Google Gemini and OpenRouter from a single endpoint. Smart auto-routing picks the best model, and automatic fallback keeps you online.
0
Providers
0
Models
Free
& Open Source
0
Unified API
Routing across leading providers
Built to keep you shipping
Everything you need to talk to any model, without vendor lock-in or downtime.
Smart Auto-Routing
A routing agent inspects each request and picks the best-fit model for speed, cost, or capability.
Automatic Fallback
If a provider is down or rate-limited, requests reroute instantly to a healthy one. Zero downtime.
Free & Paid Tiers
Start on generous free models, then upgrade to premium tiers per request — you stay in control.
Quickstart
Drop-in compatible with the OpenAI SDK
Already using OpenAI's client? Point base_url at Open LLM and
you're done. Every request is routed, load-balanced and given automatic fallback.
from openai import OpenAI
client = OpenAI(
base_url="https://open-llm-2fgm.onrender.com/v1",
api_key="YOUR_API_KEY"
)
response = client.chat.completions.create(
model="openai/...",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
Open Source
Built in the open
Open LLM is fully open source under the MIT license. Inspect every line of the router, self-host it on your own infrastructure, or contribute back to the project. We welcome bug fixes, new provider integrations, documentation improvements and feature ideas — no contribution is too small.
- Inspect the code — nothing is hidden behind a black box.
- Self-host — run the router on your own infrastructure.
- Contribute — add providers, fix bugs, improve docs.
Ready to route your first request?
Grab a free API key in seconds. No credit card, no subscription.
Get your free key