llama.cpp Server
llama.cpp is a high-performance LLM inference server with OpenAI-compatible API. This installation uses the Vulkan backend for GPU acceleration across AMD, Intel, and NVIDIA GPUs. Features automatic model downloading from HuggingFace, built-in web UI, and supports multiple concurrent requests.
Install
bash -c "$(curl -fsSL https://raw.githubusercontent.com/Heretek-AI/ProxmoxVE/main/ct/llamacpp.sh)"Run as root in the Proxmox VE shell. Pick Default for the sizing below, or Advanced to choose storage, network and resources yourself.
Notes
- Warning: Requires GPU passthrough - configure in Proxmox before starting container
- Warning: AMD GPU: Add 'dev0: /dev/kfd,gid=104' and 'dev1: /dev/dri/renderD128,gid=104' to container config
- Warning: Intel GPU: Add 'dev0: /dev/dri/renderD128,gid=104' to container config
- Warning: NVIDIA GPU: Requires nvidia-container-toolkit on host
- Access the Web UI at http://<IP>:8080
- OpenAI-Compatible API: http://<IP>:8080/v1/chat/completions
- Default model: unsloth/Qwen3.5-9B-GGUF:Q8_0 (auto-downloaded on first start)
- Models are stored in /opt/llamacpp/models/
- Change model: Edit /etc/systemd/system/llamacpp.service and restart
- GPU passthrough guide: /opt/llamacpp/GPU_PASSTHROUGH.md
What's new upstream
Fetching releases from GitHub…
Script details
You might also need
Hermes Agent
LXCAI / Coding & Dev-Tools · debian 13
Hermes Agent is a self-improving AI agent built by Nous Research with persistent memory, autonomous skill creation, computer-use (CUA) support, and multi-platform gateway integration. This enhanced Heretek build includes pre-configured passwordless sudo for the hermes user, native X11/input shared libraries for cua-driver, and Docker Engine with Buildx and Compose v2.
Kyojin
LXCAI / Coding & Dev-Tools · ubuntu 24.04
Kyojin is Yamz Labs' open source inference engine for AMD Strix Halo (gfx1151), built on ExLlamaV3 with speculative decoding, serving their own EXL3 quantised builds of Qwen3.8-Flash-Next, GLM-5.3-Flash and MiMo-V2.6-Flash (95 to 106 GB) on one 128 GB machine through an OpenAI-compatible API. The script builds the latest tagged release with AMD's gfx1151 ROCm wheels, following Kyojin's own container recipe.
GSQHalo.cpp
LXCAI / Coding & Dev-Tools · debian 13
GSQHalo.cpp is a llama.cpp fork tuned for ISTA-DASLab's GSQ-RCO quantizations of Qwen3.8-Flash-Next on AMD Strix Halo (gfx1151), with faster prefill kernels, MTP that fits in memory and a KV cache on SSD. The script installs AMD's ROCm 7.14.1 for gfx1151 in its own prefix, builds the fork from source and serves llama-server's OpenAI-compatible API.
Gufo
LXCAI / Coding & Dev-Tools · debian 13
Gufo is an open source local inference engine built and tuned for AMD Strix Halo (Ryzen AI MAX+ 395, Radeon 8060S gfx1151). It serves Qwen3.8, DeepSeek V4 Flash, speech, embedding, image and video models through an OpenAI-compatible API, with speculative decoding, concurrent requests and prompt caching. The script installs the official release build through Nix, using prebuilt ROCm from the Gufo binary cache.