llama.cpp ghcr.io/ggml-org/llama.cpp:full ghcr.io/ggml-org/llama.cpp full Latest full release light Latest light release server Latest server-only release full-cuda Latest full release with CUDA support light-cuda Latest light release with CUDA support server-cuda Latest server-only release with CUDA support full-rocm Latest full release with ROCm support light-rocm Latest light release with ROCm support server-rocm Latest server-only release with ROCm support full-musa Latest full release with MUSA support light-musa Latest light release with MUSA support server-musa Latest server-only release with MUSA support full-intel Latest full release with SYCL support light-intel Latest light release with SYCL support server-intel Latest server-only release with SYCL support full-vulkan Latest full release with Vulkan support light-vulkan Latest light release with Vulkan support server-vulkan Latest server-only release with Vulkan support bridge http://[IP]:[PORT:8000]/ -m /models/model.gguf --port 8000 --host 0.0.0.0 --n-gpu-layers 1 false https://github.com/ggml-org/llama.cpp/issues https://github.com/ggml-org/llama.cpp Inference of Meta's LLaMA model (and others) in pure C/C++ False AI: Productivity: Tools: Other: Status:Stable https://user-images.githubusercontent.com/1991296/230134379-7181e485-c521-4d23-a0d6-f7b3b61ba524.png https://user-images.githubusercontent.com/1991296/230134379-7181e485-c521-4d23-a0d6-f7b3b61ba524.png https://raw.githubusercontent.com/nwithan8/unraid_templates/main/templates/llama_cpp.xml https://github.com/nwithan8 The image for this container is several gigabytes. If you receive a "no space left on device" warning during installation, please increase the vDisk size in your Docker settings. This container expects a "model.gguf" file to be present in the model storage path. If you are using an Nvidia GPU, add "--gpus all" to the Extra Parameters field under Advanced. ### 2026-02-23 Fix repo location and branch options ### 2025-05-03 Add MUSA branch ### 2024-05-07 Initial release 8000 /mnt/user/appdata/llama_cpp/model