torch-xpu-bench SKILL.md skillA
torch-xpu-bench from intel/skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add intel/skills/torch-xpu-bench@git:20261105.847cd7aPins git:20261105.847cd7a in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- intel/skills
- Path
skills/torch-xpu-bench/SKILL.md- Versions recorded
- 2
- Latest version
- git:20261105.847cd7a, committed Nov 5, 2026
- First version recorded
- Sep 3, 2026
- Size
- 7,783 bytes
- Repository stars
- 29
- License
- Apache-2.0
Benchmark a Hugging Face model on an Intel GPU through pure PyTorch + Transformers, **single-process, no HTTP server**. Measures generate() throughput in tokens/sec, time-to-first-token, decode-step latency, and peak XPU memory. Also covers diffusion and encoder-only models via `references/non-llm-snippets.md`. Use after **model-can-it-fit** to validate predicted memory against `torch.xpu.max_memory_allocated()`.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# torch-xpu-bench
Pure-PyTorch benchmarks of HF models on XPU. Measures bare
`model.generate()` with no scheduler, no continuous batching, no
API overhead. For serving-shape numbers (TTFT/TPOT/ITL under
concurrency) use **vllm-xpu-bench**.
**Important:** This skill provides a ready-to-use benchmark script at
`{base_dir}/scripts/bench.py`. Do not write your own benchmark script;
use the provided one.
Use cases:
- TTFT and decode rate of a single sequence at a given dtype.
- A/B dtypes (bf16 vs fp16), compile on/off, model size.
- Decide whether `--enforce-eager` is worth adding in vLLM for a
problem model (graphs are on by default from v0.31.0).
- Validate **model-can-it-fit** predictions against `max_memory_allocated`.
For diffusion and encoder-only models, see
`references/non-llm-snippets.md`.
## CUDA → XPU translation
| CUDA | XPU |
|---|---|
| `torch.cuda.synchronize()` | `torch.xpu.synchronize()` |
| `torch.cuda.Event(enable_timing=True)` | `torch.xpu.Event(enable_timing=True)` |
| `torch.cuda.max_memory_allocated()` | `torch.xpu.max_memory_allocated()` |
| `torch.cuda.reset_peak_memory_stats()` | `torch.xpu.reset_peak_memory_stats()` |
…Read the whole file at its exact version.
How to install
mdr add intel/skills/torch-xpu-bench@sha256:d7a7d12ea24cd00bThe line at the top of this page pins the label git:20261105.847cd7a, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_es3y3xm275tmvl3c)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, between Sep 3, 2026 and Nov 5, 2026, about one every 63 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261105.847cd7a latest | 2026-11-05 | 847cd7a | 7,783 B | A | view · diff |
| git:20260903.b1ad034 | 2026-09-03 | b1ad034 | 7,482 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (7783 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
intel/skills · 29 stars · license Apache-2.0 · pushed 2026-11-06 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_es3y3xm275tmvl3c GET https://markdownregistry.com/api/v1/resolve?ref=intel/skills/torch-xpu-bench GET https://markdownregistry.com/api/v1/blob/d7a7d12ea24cd00bff31a721cedf2f16f0b0d4eb5629ca887b6814ca231c90c6
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.