Ooor review
A free, open-source Windows console for managing llama.cpp models, runtimes, chat, and APIs.
Reviewed by iTechGuides Editors · Editorial team · Updated Nov 2026
Ooor is a native Windows desktop workspace for running and organizing local large language models through llama.cpp. It suits solo users and small teams who want one console for installing runtime builds, downloading GGUF models, chatting locally, and exposing a localhost API. The application is open source under the MIT license, supports offline use, and stores data locally. Its interface is available in English and Simplified Chinese.
There is no paid tier in the published offering: Ooor is distributed as free, open-source software. Its core workflow covers runtime switching across CPU, CUDA, Vulkan, and SYCL builds; a local GGUF library with metadata and projector files; Hugging Face model search; and chunked, resumable downloads. Profiles can preserve engine, model, and agent configurations. For application integration, the OpenAI-compatible localhost endpoint connects local workloads to a familiar API shape, while llama.cpp, Hugging Face, and Model Context Protocol define the surrounding ecosystem.
Ooor’s breadth extends beyond chat through agent tools for URL fetching, file access, shell commands, memory, and MCP server binding. Streaming responses with token timing add visibility during local sessions. That feature set makes it a practical control console for Windows users who specifically want llama.cpp management without a subscription. The trade-off is platform scope: it is Windows-only, with desktop deployment and no stated support for other operating systems. Users seeking a broader multi-platform local-LLM manager or a runtime-agnostic workspace should consider alternatives; those comfortable with Windows and llama.cpp get a concentrated, no-cost toolkit.
Ooor pros and cons
- Where it wins
- Switches llama.cpp CPU, CUDA, Vulkan, and SYCL runtimes
- Combines GGUF model management with Hugging Face browsing
- Offers local chat, OpenAI-compatible API, agents, and MCP support
- Where it doesn't
- Windows desktop availability limits cross-platform use
- Agent shell, file, and URL tools require careful local control
- Focused on llama.cpp rather than a broader runtime ecosystem
Ooor fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update