Inference of open‑source LLMs
at the lowest possible cost

Maximum privacy & control. Your software talks directly to the Gonka network — we never see your prompts or completions.

• ✓ Direct peer-to-peer inference
• ✓ Data never passes through our servers
• ✓ Full control over endpoint selection
• ✓ Requires a few lines of integration

import OpenAI from "openai";
import { resolveAndSelectEndpoint, gonkaFetch } from "gonka-openai";

const { selected } = await resolveAndSelectEndpoint({
  sourceUrl: "https://node3.gonka.ai",
});

const fetch = gonkaFetch({
  gonkaPrivateKey: process.env.YOUR_PRIVATE_API_KEY,
  selectedEndpoint: selected,
});

const client = new OpenAI({
  apiKey: "any-string",
  baseURL: selected.url,
  fetch,
});

const response = await client.chat.completions.create({
  model: "Qwen/Qwen3-32B-FP8",
  messages: [{ role: "user", content: "Hello!" }],
});
    

Instant compatibility. Plug any OpenAI‑compatible tool into Gonka — just replace the base_url and API key.

• ✓ Works with LangChain, LlamaIndex, ChatGPT‑like UIs
• ✓ Zero code changes required
• ✓ All requests are forwarded (we do not store them)
• ✓ Ideal for existing projects and no‑code integrations

from openai import OpenAI

client = OpenAI(
    base_url="https://proxy.gonkabroker.com/v1",  
    api_key="your_broker_key"  
)

response = client.chat.completions.create(
    model="Qwen/Qwen3-32B-FP8",
    messages=[{"role": "user", "content": "Hello!"}]
)