Ollama, the open source LLM server, exposing a private GPU accelerated OpenAI compatible API within minutes of launch.
This product is available in the build below. Open it for the operating system, cloud and version detail, or read its deployment guide.
| Product | Cloud | Operating system | Version | |
|---|---|---|---|---|
| Ollama GPU-Accelerated LLM Server | AWS | — | Standard | View · Guide |
Ollama is an open source server for running large language models locally. This image exposes an OpenAI compatible REST API behind an nginx proxy and pre pulls a starter model so the endpoint responds immediately. It suits private inference for models such as Llama, Mistral, Gemma and Qwen.
The cloudimg image ships Ollama as a system service with the NVIDIA datacenter driver preinstalled and verified on real hardware, so GPU inference works out of the box on supported instances. HTTP Basic Authentication gates every sensitive endpoint with a unique password generated per instance on first boot, model weights live on a dedicated disk, and no default credentials ship. Backed by 24/7 cloudimg support for GPU sizing, model selection and scaling.