OpenAI-compatible · Ollama-powered

Plant a prompt.
Watch it grow
into code.

A self-serve LLM inference API with API keys, usage analytics, and pay-per-token pricing. Drop in your baseURL and go.

Get an API keyRead the docs

Free to start · No credit card required

Available models: Llama 3.1, Llama 3.1, Qwen 2.5, Qwen 2.5, Mistral, Deepseek Coder, Phi 3, Gemma 2

Quickstart

From zero to inference
in three steps.

If you already use the OpenAI SDK, you're 90% done. Change two lines, keep the rest.

01

Get an API key

Create a free account, then generate a key from the dashboard. Keys are prefixed soc_live_ so they're easy to spot in your code.

# From the dashboard → API Keys → Create key
# Copy the key — you only see it once
02

Point your client at SeedofCode AI

It's one line. SeedofCode AI is OpenAI-API-compatible, so any SDK that supports a custom baseURL works — Python, Node, Go, or plain curl.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.ai.seedofcode.dev/api",
  apiKey: "soc_live_your_key_here",
});
03

Call the API

Use the same chat/completions interface you already know. Pick any available model — llama, qwen, mistral, and more.

const response = await client.chat.completions.create({
  model: "qwen2.5vl:7b",
  messages: [
    { role: "user", content: "Plant a prompt." }
  ],
});
console.log(response.choices[0].message.content);
// → "Watch it grow into code."

Why SeedofCode AI

LLM inference without the noise.

OpenAI-compatible

Drop in your baseURL — the same SDKs, same request shapes, same response format.

Pay per token

Credits deducted per request. No monthly subscriptions, no seat limits, no surprises.

Real models, real hardware

Running Llama, Qwen, Mistral and more on dedicated GPU nodes via Ollama — not a wrapper.

Zero Data Retention

We never store your prompts or completions. Your data is processed in memory and instantly discarded.

Documentation

Everything you need to build.

Browse all docs

Quickstart

Get a key, set your baseURL, make your first request. Five minutes, working code.

Read more

API Keys

Create, rotate and revoke keys. Understand the soc_live_ / soc_test_ prefix system.

Read more

SDK Compatibility

Drop-in OpenAI-compatible. Works with the official SDKs for Python, Node, Go and more.

Read more