Introduction

Psyllama provides a local HTTP API for chat, text generation, embeddings, and model management. The API is designed to be easy to call from scripts and applications.

Get started

Install Psyllama and run a model first:

curl -fsSL https://psyllama.com/install.sh | sh
psyllama run kimi-k2.5:cloud

Base URL

By default, Psyllama listens on:

http://localhost:11434/api

Example request

curl http://localhost:11434/api/chat \
  -d '{
    "model": "kimi-k2.5:cloud",
    "messages": [
      {"role": "user", "content": "Why is the sky blue?"}
    ]
  }'

Authentication

For local usage, no authentication is required. If you expose Psyllama over a network, put it behind an authenticating reverse proxy or firewall rules.

Versioning

CLI and server versions are reported by the version endpoint.

curl http://localhost:11434/api/version

Back to docs →