README.md

---
title: Ollama Server
emoji: 🦙
colorFrom: blue
colorTo: green
sdk: docker
app_port: 11434
pinned: true
---

# Ollama Server on Hugging Face Spaces

A lightweight [Ollama](https://ollama.com/) inference server running on a Hugging Face Space.

## Pre-pulled Models

- `gemma2:2b`
- `alibilge/Huihui-GLM-4.6V-Flash-abliterated:q4_k_s`
- `frob/mradermacher-Llama3.3-8B-Thinking-Heretic-Claude-4.5-Opus:q8_0`

## Usage

Once the Space is running, you can interact with the Ollama API at the Space's URL.

### Pull additional models

```bash
curl https://your-space.hf.space/api/pull -d '{"model": "llama3.2"}'
```

### Generate a completion

```bash
curl https://your-space.hf.space/api/generate -d '{
  "model": "frob/mradermacher-Llama3.3-8B-Thinking-Heretic-Claude-4.5-Opus:q8_0",
  "prompt": "Hello!"
}'
```

## API

This Space exposes the standard Ollama REST API on port `11434`.