Intel GPU AI Skills is a collection of agent skills for setting up, running, benchmarking, and profiling Hugging Face models on Intel GPUs. It supports workflows involving PyTorch, vLLM-XPU, SGLang-XPU, llama.cpp-SYCL, and migration from CUDA to XPU. The catalogue contains the project's skills, instructions, agent, and plugin.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add intel/gpu-ai-skills --skill llamacpp-xpu-rungit clone --depth 1 https://github.com/intel/gpu-ai-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/intel/gpu-ai-skills/llamacpp-xpu-run)<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/llamacpp-xpu-run"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/llamacpp-xpu-run/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/llamacpp-xpu-run"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/llamacpp-xpu-run.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 8 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 27 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 61 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 89 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 91 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- medium MCP Rug Pull · line 60 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium MCP Rug Pull · line 76 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium MCP Rug Pull · line 87 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium Data Exfiltration · line 120 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00163 | $0.01583 |
| Opus 5 | $0.00081 | $0.00792 |
| Sonnet 5 | $0.00033 | $0.00317 |
| Haiku 4.5 | $0.00016 | $0.00158 |
Grade A, and why
llamacpp-xpu-run scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -f http://localhost:8000/health # {"status":"ok"} How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
llamacpp-xpu-run
Run GGUF models on Intel GPUs via llama.cpp's SYCL backend.
The official .devops/intel.Dockerfile is the canonical build path.
Tested against tag b9494 on Intel Arc B70 (Battlemage, level_zero:0).
CRITICAL SAFETY RULE: When removing a docker container, always docker stop first, then docker rm. Never use docker prune or any system-wide process-kill command.
Performance note
When the Docker image and GGUF model are already present, launch the server immediately. Skip redundant image rebuilds and device checks — the server will fail fast if misconfigured. Always verify with the health endpoint after launch (see Validate section below).
Use a pinned upstream tag (e.g. :b9494), not :latest. docker run reuses any locally-tagged image with no upstream check, so a stale llama-server-sycl:latest will silently run. If the user requests :latest, rebuild with --no-cache first, or resolve latest to the current upstream tag and build with that pinned name.
CUDA → XPU cheat sheet
| CUDA | Intel SYCL |
|---|---|
-DGGML_CUDA=ON |
-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx |
CUDA_VISIBLE_DEVICES=0 |
ONEAPI_DEVICE_SELECTOR="level_zero:0" |
--gpus all (Docker) |
--device /dev/dri --group-add render |
--n-gpu-layers 99 |
same (-ngl 99) |
Before running any docker commands below, confirm with the user that they want to proceed. The following steps will launch a Docker container and may modify system state.
Build
Confirm the tag and GGUF model with the user before running — the build pulls source and creates a Docker image.
git clone --depth 1 --branch <tag> https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
docker build \
--build-arg http_proxy=$http_proxy \
--build-arg https_proxy=$https_proxy \
--build-arg GGML_SYCL_F16=OFF \
--target server \
-t llama-server-sycl:<tag> \
-f .devops/intel.Dockerfile .
Three targets available — server (serving), full (bench + convert), light
(cli + batch). See references/build-and-env.md for proxy setup, the
Level Zero deb conflict fix, and full env var reference.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 144 lines · 163 tokens per session scan A 7963fde6360e
llamacpp-xpu-run is a skill published in the GitHub repository intel/gpu-ai-skills (21 stars, last pushed 5d ago), licensed Apache-2.0. It adds 163 tokens to every session and 1,583 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-memory-thermal-ops
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
llama-cpp
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
minicpm5-deploy-vllm-ascend
Deploy MiniCPM5-2B with vLLM on Huawei Ascend NPU using vLLM-Ascend. Use when the user mentions vLLM-Ascend, Ascend NPU, Huawei Ascend, CANN, torchnpu, davinci devices, or wants an OpenAI-compatible MiniCPM5 server on Ascend hardware.
amc-run-video-calibration
Calibrates pre-recorded cam.mp4 datasets through the AutoMagicCalib REST API. Use for user-supplied local MP4s; route live RTSP streams to amc-run-rtsp-calibration.