Ollama on Coiled

chat.py

import coiled
import ollama
from distributed import print

@coiled.function(name="ollama")
def chat():
    model = "llama3:latest"
    print(f"Pulling {model} model...")
    ollama.pull(model)
    print(f"Done pulling {model}")

    stream = ollama.chat(
        model=model,
        messages=[{"role": "user", "content": "Why is the sky blue?"}],
        stream=True,
    )
    for chunk in stream:
        print(chunk["message"]["content"], end="", flush=True)

chat()

serve.sh

#!/usr/bin/env bash

curl https://ollama.ai/install.sh | sh
export OLLAMA_HOST="0.0.0.0"
ollama serve