vllm-quantization SKILL.md skillA
vllm-quantization 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-quantization@git:20261018.1fb9d2dPins git:20261018.1fb9d2d 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-quantization/SKILL.md- Versions recorded
- 12
- Latest version
- git:20261018.1fb9d2d, committed Oct 18, 2026
- First version recorded
- Aug 18, 2026
- Size
- 26,782 bytes
- Repository stars
- 6
- License
- MIT
vLLM datacenter-GPU quantization — picking, configuring, troubleshooting NVFP4, FP8, MXFP4, MXFP8, AWQ, GPTQ, INT8, compressed-tensors, modelopt, quark on H100/H200/B200/B300/GB200/GB300; every `--quantization` flag value, KV-cache dtypes (fp8_e4m3, nvfp4, per-token-head, turboquant), MoE backend selection (CUTLASS, TRTLLM, FlashInfer, DeepGEMM, Marlin, Qutlass), producing checkpoints with llm-compressor and NVIDIA ModelOpt (NVFP4_DEFAULT_CFG, FP8_DEFAULT_CFG, W4A16, SmoothQuant+GPTQ), online…
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 quantization — operator skill Registries, counts and flag values below are re-derived from the **v0.32.0** tree (2026-10-18); verification dates and citations live in `references/sources.md`. For production vLLM operators on **H100 / H200 / B200 / B300 / GB200 / GB300** fleets deciding which quantization format fits a given target model, producing a checkpoint vLLM will actually load, wiring the right KV-cache dtype, diagnosing accuracy or throughput regressions after an upgrade, and composing quantization with speculative decoding / LoRA / MoE. Pointer-map format: this SKILL.md picks the format and CLI; the files in `references/` hold the per-format deep dives, exact source pointers, and troubleshooting cards. Follow the link, don't paraphrase from memory — the quantization layer moves faster than any other subsystem in vLLM (six formats landed in v0.19 alone). ## When quantization wins, when it doesn't Quantization trades weight precision for memory + compute: - **KV-capacity bound** (long context, high concurrency) — FP8 or NVFP4 **KV cache** gives a 2×/4× KV-capacity multiplier; weight format matters much …
Read the whole file at its exact version.
How to install
mdr add air-gapped/skills/vllm-quantization@sha256:0370272b8c52ede8The line at the top of this page pins the label git:20261018.1fb9d2d, 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_fcsjwhjrql74f7yh)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 12 versions of this file, between Aug 18, 2026 and Oct 18, 2026, about one every 6 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261018.1fb9d2d latest | 2026-10-18 | 1fb9d2d | 26,782 B | A | view · diff |
| git:20261018.d9dcd1c | 2026-10-18 | d9dcd1c | 27,335 B | A | view · diff |
| git:20261018.941f07d | 2026-10-18 | 941f07d | 27,558 B | A | view · diff |
| git:20261018.a384679 | 2026-10-18 | a384679 | 31,151 B | A | view · diff |
| git:20261018.41dec7b | 2026-10-18 | 41dec7b | 31,464 B | A | view · diff |
| git:20260923.466430c | 2026-09-23 | 466430c | 27,884 B | A | view · diff |
| git:20260922.fd88423 | 2026-09-22 | fd88423 | 26,203 B | A | view · diff |
| git:20260915.23ffb18 | 2026-09-15 | 23ffb18 | 25,886 B | A | view · diff |
| git:20260915.1702bca | 2026-09-15 | 1702bca | 25,797 B | A | view · diff |
| git:20260915.9e1fc69 | 2026-09-15 | 9e1fc69 | 25,414 B | A | view · diff |
| git:20260915.34767df | 2026-09-15 | 34767df | 24,390 B | A | view · diff |
| git:20260818.6de1cf4 | 2026-08-18 | 6de1cf4 | 24,175 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 (26782 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_fcsjwhjrql74f7yh GET https://markdownregistry.com/api/v1/resolve?ref=air-gapped/skills/vllm-quantization GET https://markdownregistry.com/api/v1/blob/0370272b8c52ede87a18278861700b1b5c13ffb08291627020c2c8c3e80b5b70
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.