Quickstart
Two steps: stand up a worker to run inference, then call the API. The
Pendra API is OpenAI-compatible, so if you've used the OpenAI SDK before,
the shapes will look familiar — just point at https://api.pendra.ai
with a Pendra API key.
1. Set up a worker
Every Pendra request runs on a worker — either GPU capacity we provision for you, or a worker you run on your own hardware. There's no shared pool of hosted models to call, so the worker comes first. Pick one to get started.
Pendra-managed
Dedicated GPU capacity provisioned on Pendra-managed infrastructure in the UK. Zero ops, same API, available on the Enterprise plan. Get in touch and we'll size capacity to your workload.
Get in touchSelf-hosted
Install the Pendra worker on your own GPUs — it serves chat, embeddings, image generation, and audio transcription out of the box. Inference runs only on your hardware.
Install a worker2. Get an API key
Don't have an account yet? Sign up at console.pendra.ai — see pricing for the available plans. The first time you sign in, we ask what brought you to Pendra and how you want to run it, and take you to the right place to start. Both questions are optional, and you can skip them.
Once signed in, open API Keys and create a new key. Keys
are shown once at creation — copy and store it now. The key looks like
pdr_sk_….
3. Install an SDK
Or skip straight to curl below if you don't need a client library.
Python
pip install pendra
Node
npm install pendra
Prefer another language? The OpenAI and Anthropic SDKs work too — just point the base URL at https://api.pendra.ai.
4. Make your first request
Python
from pendra import Pendra
client = Pendra(api_key="pdr_sk_...")
response = client.chat.completions.create(
model="qwen3.6:27b",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
Node
import { Pendra } from "pendra";
const client = new Pendra({ apiKey: "pdr_sk_..." });
const response = await client.chat.completions.create({
model: "qwen3.6:27b",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);
curl
curl https://api.pendra.ai/api/v1/chat/completions \
-H "Authorization: Bearer pdr_sk_..." \
-H "Content-Type: application/json" \
-d '{"model":"qwen3.6:27b","messages":[{"role":"user","content":"Hello"}]}'
Next steps
- Browse the full API reference for embeddings, image generation, audio transcription, and the Anthropic-compatible Messages API.
- Point a coding agent at Pendra: Claude Code, Codex, OpenClaw.
- Need full data residency? Run a self-hosted worker.