vllm-nvidia-hardware SKILL.md skillA
vllm-nvidia-hardware from air-gapped/skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add air-gapped/skills/vllm-nvidia-hardware@git:20261018.62c3321Pins git:20261018.62c3321 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- air-gapped/skills
- Path
.claude/skills/vllm-nvidia-hardware/SKILL.md- Versions recorded
- 10
- Latest version
- git:20261018.62c3321, committed Oct 18, 2026
- First version recorded
- Apr 19, 2026
- Size
- 9,559 bytes
- Repository stars
- 6
- License
- MIT
NVIDIA AI-hardware + vLLM-platform reference covering Hopper (H100/H200), Blackwell (B100/B200/B300) and Blackwell Ultra, Grace-Blackwell superchips and NVL72 racks (GB200, GB300), Vera Rubin (R100/R300) with VR200 NVL144 and Kyber NVL576, Dell PowerEdge XE family and IR5000/IR7000/IR9048 racks. Per-SKU HBM, FP4/FP8/FP16 TFLOPs, NVLink5, TDP, rack power/cooling (135 kW GB300, 180-220 kW NVL144, 600 kW Kyber), DLC vs RDHx, 800 VDC HVDC. Memory-wall roofline, HBM3E→HBM4 supply 2026. vLLM…
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
# vLLM on NVIDIA hardware — Hopper through Rubin
Answer from the shortcuts below; for full tables, read the one reference file the router names for the question.
## The one thing to know before anything else
LLM inference has two phases with radically different bottlenecks:
- **Prefill** is compute-bound (GEMMs, AI ≫ ridge point) — more FLOPs help.
- **Decode** is memory-bandwidth-bound (AI ≈ 1, 100× below the ridge) — more HBM bandwidth helps, more FLOPs don't.
Every hardware decision — FP4 vs FP8, B300's higher FLOPs with the same 8 TB/s, NVL72's domain collapse, Rubin's HBM4 jump to ~20 TB/s — is about relieving the memory wall on decode while keeping prefill healthy. Read `references/fundamentals.md` for the roofline math and the HBM roadmap context that makes the rest of the tables meaningful.
## Quick-answer router
**Hardware specs** ("what's the HBM on X?", "TDP of Y?")
- NVIDIA GPU SKUs (Hopper, Blackwell, Blackwell Ultra) → `references/gpu-specs.md`
- Vera Rubin roadmap (R100, Rubin Ultra, NVL144, Kyber NVL576) → `references/rubin-roadmap.md`
- Dell PowerEdge XE servers → `references/dell-xe.md`
…Read the whole file at its exact version.
How to install
mdr add air-gapped/skills/vllm-nvidia-hardware@sha256:c9c117b765e7d96cThe line at the top of this page pins the label git:20261018.62c3321, 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_hug6vmcchvtf3m2g)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 10 versions of this file, between Apr 19, 2026 and Oct 18, 2026, about one every 20 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261018.62c3321 latest | 2026-10-18 | 62c3321 | 9,559 B | A | view · diff |
| git:20261018.bb4560b | 2026-10-18 | bb4560b | 9,510 B | A | view · diff |
| git:20261018.6e49468 | 2026-10-18 | 6e49468 | 9,495 B | A | view · diff |
| git:20261018.b9d519e | 2026-10-18 | b9d519e | 10,260 B | A | view · diff |
| git:20261018.fa3e7b2 | 2026-10-18 | fa3e7b2 | 11,500 B | A | view · diff |
| git:20260811.6e4da75 | 2026-08-11 | 6e4da75 | 11,838 B | A | view · diff |
| git:20260721.b403624 | 2026-07-21 | b403624 | 10,072 B | A | view · diff |
| git:20260529.401d93d | 2026-05-29 | 401d93d | 8,277 B | A | view · diff |
| git:20260424.4dca59c | 2026-04-24 | 4dca59c | 8,362 B | A | view · diff |
| git:20260419.ef90b92 | 2026-04-19 | ef90b92 | 8,293 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 (9559 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
air-gapped/skills · 6 stars · license MIT · pushed 2026-11-04 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_hug6vmcchvtf3m2g GET https://markdownregistry.com/api/v1/resolve?ref=air-gapped/skills/vllm-nvidia-hardware GET https://markdownregistry.com/api/v1/blob/c9c117b765e7d96c6b392729a53e8169d66d6c4a070cc1e330cbb2bf91423477
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.