The Ollama model driver for Namzu.

Install · Usage · Documentation

Implements the kernel's LLMProvider interface over a local Ollama server, so a run can be driven by a model on your own machine with no key and no egress. Installed only if you use it — the kernel has no preferred vendor and no driver is a dependency of it.

Install

pnpm add @namzu/sdk @namzu/ollama

@namzu/sdk is a peer dependency. Install both.

Usage

import { ProviderRegistry } from '@namzu/sdk'
import { registerOllama } from '@namzu/ollama'

// Once, at startup.
registerOllama()

const { provider } = ProviderRegistry.create({
  type: 'ollama',
  model: 'llama3.2', // default; overridable per call
})

for await (const chunk of provider.chatStream({
  model: 'llama3.2',
  messages: [{ role: 'user', content: 'Hello' }],
})) {
  if (chunk.delta.content) process.stdout.write(chunk.delta.content)
}

chatStream is the only model entry point; a non-streaming call is that stream collected. In practice the kernel's run loop calls it and hands you events.

Documentation

• The Ollama driver — configuration, refusals and cancellation
• Namzu docs

License

FSL-1.1-MIT, converting to MIT two years after each release.