One key. Every model. The cheapest healthy seller.
UberLLM is an OpenAI-compatible LLM gateway across 251 models that routes every request to the cheapest healthy seller on a live marketplace and bills the price shown, fee included. Switching from OpenRouter takes one line — change the base URL.
Migrate in one line
from openai import OpenAI
client = OpenAI(
base_url="https://api.uberllm.dev/v1", # was: https://openrouter.ai/api/v1
api_key="ull_your_key_here",
)
resp = client.chat.completions.create(
model="deepseek/deepseek-v4-flash",
messages=[{"role": "user", "content": "Hello!"}],
)UberLLM vs OpenRouter
| Capability | UberLLM | OpenRouter |
|---|---|---|
| OpenAI-compatible, one key | Yes | Yes |
Comparison reflects publicly documented features; both are independent products. OpenRouter is a trademark of its owner.
FAQ
Is UberLLM a drop-in OpenRouter alternative?
Yes. UberLLM exposes one OpenAI-compatible base URL, so switching from OpenRouter (or OpenAI) means changing only the base_url and api_key — your existing SDK code stays the same. Anthropic SDKs work via /v1/messages.
How is UberLLM different from OpenRouter?
Both give you one key for many models. UberLLM routes every request to the cheapest healthy seller for that model and bills you the price shown, which includes UberLLM's 10% fee; a 5% fee applies when you buy credits. It also publishes a live board of per-model prices.
Does UberLLM support the same models?
UberLLM covers GPT, Claude, Gemini, DeepSeek, GLM, Qwen, Llama, Mistral, Kimi and more — one key across all of them. See the live model list for exactly what you can call today.
How do I migrate from OpenRouter to UberLLM?
Set base_url to https://api.uberllm.dev/v1 and use your UberLLM key. Model ids follow the familiar provider/model form. That is the whole migration.