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 xpu-model-type-detectgit 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/xpu-model-type-detect)<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/xpu-model-type-detect"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/xpu-model-type-detect.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00141 | $0.02144 |
| Opus 5 | $0.00071 | $0.01072 |
| Sonnet 5 | $0.00028 | $0.00429 |
| Haiku 4.5 | $0.00014 | $0.00214 |
Grade A, and why
xpu-model-type-detect scanned grade A with 0 findings 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 8d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
xpu-model-type-detect
Pick the right loader class before you load. A wrong loader class produces an opaque failure after the model is already on XPU — 20+ seconds into a load you didn't need to make.
Reporting the verdict
Run scripts/detect.py against the model id rather than answering
its type from memory — even for a familiar model, the detector's
verdict is the citable result. Then paste its output block verbatim
in a fenced code block, keeping the detected:, loader:, inputs:,
and confidence: lines as-is rather than reformatting them into a
table — those literal labels are what the user and downstream tools
key on. Lead with the confirmed type, e.g. "detection confirmed:
multimodal_vl", to show the verdict came from the run.
Detection is read-only, so there's nothing to health-check afterward — confirming the verdict block is the result.
Quickstart
python3 scripts/detect.py --model openai/clip-vit-base-patch32
model_id: openai/clip-vit-base-patch32
detected: vision_language
loader: transformers.CLIPModel
processor: transformers.AutoProcessor
inputs: pixel_values, input_ids
rationale: architectures[0]=CLIPModel; text + vision towers detected
Stdlib only. HF_TOKEN needed for gated repos.
With --json the output is a single JSON object suitable for piping
into another tool.
3-stage detection
The script tries three signals in order; first hit wins.
- Name pattern on
model_id— cheapest. Currently catches reward models (*-rm,*-reward), embedding repos (*-embed*,bge-,gte-,e5-,jina-embed*), and incompatible checkpoints (*-mlx*,*-gguf*). config.architectures[0]— authoritative when present. Pulled directly fromconfig.jsonon the Hub (no weight download). This is where most non-name-matched models classify.- HF Hub
pipeline_tag— last resort; one HTTP call to/api/models/<id>.
If all three miss, the script prints unknown and suggests reading the
model card's first usage snippet.
What ships with it
1 file 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.
- 8d ago First seen · 171 lines · 141 tokens per session scan A be03424c6dc9
xpu-model-type-detect is a skill published in the GitHub repository intel/gpu-ai-skills (21 stars, last pushed 3d ago), licensed Apache-2.0. It adds 141 tokens to every session and 2,144 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. 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.
amc-run-rtsp-calibration
Calibrate a new dataset from live RTSP camera streams via the AutoMagicCalib REST API. Use when the user provides RTSP URLs or asks to calibrate live cameras; VIOS records clips, AMC ingests them, then runs calibration.
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.