Build on Foundry

Inference a co-owned model in three lines.

The Foundry SDK ships with adapters for the agent frameworks your project already uses. Revenue routes back to the Ingot’s co-owners automatically, on-chain.

Adapters

import { foundry } from '@foundryprotocol/sdk/adapters/vercel-ai';
import { generateText } from 'ai';

const model = foundry('ingot:0x8e2…f4a');
const { text } = await generateText({
  model,
  prompt: 'Translate to Konkani: …',
});

import { FoundryChat } from '@foundryprotocol/sdk/adapters/langchain';

const llm = new FoundryChat({ ingotId: '0x8e2…f4a' });
const out = await llm.invoke('Translate to Konkani: …');

// any tool that speaks OpenAI's API
fetch('https://api.foundryprotocol.xyz/v1/chat/completions', {
  method: 'POST',
  headers: {
    'content-type': 'application/json',
    'x-foundry-ingot-id': '0x8e2…f4a',
  },
  body: JSON.stringify({ messages: [{ role: 'user', content: '…' }] }),
});