# Ollama Intel GPU

Ollama Intel GPU is a self-hosted ai app. Ollama for Intel Arc GPUs (B580, A770, A750, etc.) powered by Intel IPEX-LLM. It is open source and self-hosted, installable on Docker Hub and Unraid.

Canonical page: https://ourlocalhost.com/app/2057/ollama-intel-gpu

## Description

Ollama for Intel Arc GPUs (B580, A770, A750, etc.) powered by Intel IPEX-LLM. Drop-in replacement for the standard Ollama container — exposes the same API on port 11434. Requires an Intel Arc GPU

## Platform availability

Available on 1 of 10 platforms: Unraid.

## Health score

42/100 (Fair), based on 5 of 5 factors.
- Maintained: 50 (last commit 231 days ago)
- Popular: 27 (22 GitHub stars)
- Easy to install: 35 (1 install platform, install notes, Docker image)
- Light to run: 10 (16 GB minimum RAM)
- Security check: 75 (no serious findings)

## Links

- Docker Hub: https://hub.docker.com/r/spaceinvaderone/ollama-intel-gpu/
- GitHub: https://github.com/SpaceinvaderOne/ollama-intel-gpu
- Unraid: https://ca.unraid.net/apps/ollama-intel-gpu-02d26nm02fbw46?q=ollama-intel-gpu

## Common questions

### How can I confirm the container is using the Intel Arc GPU rather than the CPU?

Run the container with the /dev/dri device passed through and inspect the container logs. You should see oneAPI device name: Intel(R) Graphics and library=oneapi in the output.

### Where should downloaded Ollama models be stored?

Mount persistent storage to /root/.ollama when using Docker. In Unraid, use the template's Model Storage setting, which defaults to /mnt/user/appdata/ollama-intel-gpu.

### How do I connect Open WebUI to this Ollama container?

Point Open WebUI's OLLAMA_BASE_URL at the container's host IP and port 11434. Set OLLAMA_ORIGINS to * so Open WebUI is allowed to make cross-origin requests.

### What environment variables control parallelism, context, and model unloading?

OLLAMA_NUM_PARALLEL should stay at 1 for 12 GB cards. OLLAMA_NUM_CTX sets the context window in tokens, and OLLAMA_KEEP_ALIVE sets how long the model remains in VRAM, with -1 keeping it loaded indefinitely.

### How can I test the Ollama API?

Query the root health check, /api/tags, or /api/generate on port 11434. You can use curl or the container console to verify that the API is responding.

## Alternatives

- Agora: Agora is a digital shopfront that allows you to sell files directly to others.
- Alpaca - Ollama Client: Alpaca is an Ollama client where you can manage and chat with multiple models, Alpaca provides an easy and begginer friendly way of interac…
- AnythingLLM: AnythingLLM is a local, all-in-one AI app for self-hosters that works with any large language model, providing document chat, retrieval-aug…
- Ensu: Private offline on-device local LLM app by Ente, open source code, no accounts or tracking, and works cross-platform on mobile and desktop…
- Google AI Edge Gallery: A gallery that showcases on-device ML/GenAI use cases and allows people to try and use models locally.
- Jan.ai: Open-source offline AI chat software supporting local LLMs, cloud model connections, custom assistant creation, OpenAI-compatible API, and…
- LM Studio: LM Studio is an application for running and experimenting with local large language models (LLMs) such as gpt-oss, Llama, Gemma, Qwen, and…
- LocalAI: LocalAI is a free, open-source, self-hosted AI engine that provides a drop-in, OpenAI-compatible API for running local models.
- Ollama: Ollama is a local runtime for large language models (LLMs) that provides an application programming interface (API) and command-line interf…
- PocketPal: An app that brings language models directly to your phone.
- Private GPT: Ask questions to your documents without an internet connection, using the power of LLMs.
- lemonade-server: Lemonade helps users discover and run local AI apps by serving optimized LLMs right from their own GPUs and NPUs.

---
Source: Our Local Host (https://ourlocalhost.com/app/2057/ollama-intel-gpu), the self-hosted and open-source app directory.
