Skip to content
Ryvion

Quick Start

Get your first Ryvion API call working in under two minutes.

1. Get an API key

Sign up at ryvion.com and create an API key from the API Keys page. Free tier included -- no credit card required.

2. Make your first request

Ryvion is OpenAI-compatible. Use any OpenAI SDK by changing the base URL and API key.

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.ryvion.ai/v1",
    api_key="YOUR_KEY",
)

response = client.chat.completions.create(
    model="phi-4",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Node.js

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.ryvion.ai/v1",
  apiKey: "YOUR_KEY",
});

const res = await client.chat.completions.create({
  model: "phi-4",
  messages: [{ role: "user", content: "Hello" }],
});
console.log(res.choices[0].message.content);

curl

curl -X POST https://api.ryvion.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"phi-4","messages":[{"role":"user","content":"Hello"}]}'

3. Enable streaming

Add stream=True to get real-time Server-Sent Events:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.ryvion.ai/v1",
    api_key="YOUR_KEY",
)

stream = client.chat.completions.create(
    model="phi-4",
    messages=[{"role": "user", "content": "Hello"}],
    stream=True,
)

for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

4. Available models

ModelTypeNotes
phi-4ChatMicrosoft Phi-4, streaming supported
ryvion-llama-3.2-3bChatLlama 3.2 3B, streaming supported
tinyllamaChatBitNet 1-bit quantized, runs on CPU
nomic-embed-text-v1.5Embeddings768-dimensional text embeddings
sdxl-turboImageFast image generation
whisper-baseAudioSpeech-to-text transcription

List models programmatically:

curl https://api.ryvion.ai/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

5. What happens under the hood

  1. Your request hits the Ryvion hub
  2. The scheduler routes it to a verified GPU node (respecting jurisdiction preferences)
  3. The node runs inference and returns results
  4. A cryptographic receipt is generated -- Ed25519 signed proof of execution

Every response is backed by a verifiable receipt. See Cryptographic Receipts for details.

Next steps