loading models…

One endpoint. Every model.

Point your OpenAI client at zedmux and route to any provider. Usage, credits, and rate limits handled. No SDK swap, no lock-in.

…
available
1
base_url
OpenAI
+ Anthropic API

Change one line.

zedmux speaks the OpenAI API. Swap base_url and every model in the catalog is reachable through the same key.

1from openai import OpenAI
2
3client = OpenAI(
4 base_url="https://api.zedmux.tech/v1",
5 api_key="sk-zm-…"
6)
7
8r = client.chat.completions.create(
9 model="zm/model",
10 messages=[{"role":"user", "content":"Hi"}],
11)
/Anthropic/OpenAI/Google/DeepSeek/xAI/Mistral/Meta Llama/Moonshot/Zhipu/Anthropic/OpenAI/Google/DeepSeek/xAI/Mistral/Meta Llama/Moonshot/Zhipu

One API, every provider

OpenAI and Anthropic compatible on the same base_url. Route to any provider without touching your code.

Transparent quota

Every request logged with real token counts. Credits, rate limits, and usage visible from the dashboard.

One catalog, one key

The whole model catalog behind a single sk-zm- key. Add or switch models without new accounts.

live routing
GET /v1/models…

Live in four lines.

From key to first routed completion. No dashboard tour required.

zsh — zedmux
$curl https://api.zedmux.tech/v1/models
27 models · 9 providers · 1 endpoint
$export ZEDMUX_KEY=sk-zm-…
key set
$client.base_url = "https://api.zedmux.tech/v1"
drop-in OpenAI-compatible
$client.chat.completions.create(model="zm/…")
200 OK · routed · 412ms

Choose the Plan That Fits You Best

Access the best AI models through one endpoint with transparent pricing and reliable performance.

Free tokens reset daily at 07.00 WIB; paid plans reset at purchase timeSee live plans → →

Frequently asked questions

Yes. zedmux is a drop-in replacement. Change the API base URL to https://api.zedmux.tech/v1 and use your zedmux key. OpenAI Python/Node SDKs, Anthropic SDK, and cURL all work out of the box.