Developer API

Reach Angerona from your own code through the standard OpenAI SDK, with keys tied to your wallet, usage billing and a free daily allowance for holders.

The API speaks OpenAI. Aim any OpenAI SDK or tool at https://api.angerona.chat/v1, swap in an Angerona key, and chat completions, streaming, images and video run with no code changes.

• Keys tied to a wallet: sign in on angerona.chat and make them there, with no email or account step. See API keys.
• Usage billing: the owning wallet covers each call, holders' free credits first and paid credits after. See API billing.
• Anonymous mode: a secondary compute market of trusted providers serves requests and never learns the sender, though it reads the prompt. TEE stays in the app, outside API keys. See API models.

A first answer starts with a key

On angerona.chat, open Developers, sign in and press Create key. You get something like ang_live_…, displayed one time, so save it right away.

export ANGERONA_API_KEY="ang_live_..."

Call it with the OpenAI SDK

JavaScript or TypeScript:

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.ANGERONA_API_KEY,
  baseURL: "https://api.angerona.chat/v1",
});

const r = await client.chat.completions.create({
  model: "deepseek-v4.1-flash",
  messages: [{ role: "user", content: "Write a haiku about silence." }],
});

console.log(r.choices[0].message.content);

Python:

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["ANGERONA_API_KEY"], base_url="https://api.angerona.chat/v1")

r = client.chat.completions.create(
    model="claude-sonnet-5.5",
    messages=[{"role": "user", "content": "Explain zero-knowledge proofs in two sentences."}],
)

print(r.choices[0].message.content)

curl:

curl https://api.angerona.chat/v1/chat/completions \
  -H "Authorization: Bearer $ANGERONA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Hello!"}]}'

Stream the answer

const stream = await client.chat.completions.create({
  model: "claude-sonnet-5.5",
  stream: true,
  messages: [{ role: "user", content: "Tell me a short story." }],
});

for await (const chunk of stream) process.stdout.write(chunk.choices[0]?.delta?.content ?? "");

The final chunk reports usage: prompt tokens and completion tokens.

Images and video use the same client

const img = await client.images.generate({ model: "seedream-4.5", prompt: "a lighthouse at dusk, oil painting", size: "1024x1024" });
console.log(img.data[0].url); // temporary link, valid about 20 minutes

const job = await client.videos.create({ model: "your-video-model-id", prompt: "a paper boat on a rainy street", seconds: "6" });
// poll client.videos.retrieve(job.id) until status is "completed", then client.videos.downloadContent(job.id)

A safety check screens media prompts ahead of generation and refuses clearly illegal content. Video is a beta feature.

Retrying images safely: put an id of your choice in the X-Angerona-Request-Id header (8–64 characters of letters, digits, _ or -), for example client.images.generate(params, { headers: { "X-Angerona-Request-Id": id } }). After a dropped connection, GET /media/by-request/{id} hands back the images, and sending the same id again can't generate or bill a second time. Billing happens at delivery, and an undelivered image refunds itself.

The API has eight endpoints

The API keeps as little as the app

• A hash is all we hold of a key. Request logs, prompts, replies and IP addresses never get saved.
• A key's record is its name, creation date, optional spend cap and spend so far this month.
• Generated files lose GPS and device metadata and keep AI-provenance labels.

More on Data handling.

Read these next

• API keys and limits
• API billing
• API models and modes
• API errors and rate limits
• @angerona/sdk