pollard-weights SKILL.md skillA

pollard-weights from westwaters/pollard-weights is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add westwaters/pollard-weights/pollard-weights@v0.4

Pins v0.4 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
westwaters/pollard-weights
Path
skills/pollard/SKILL.md
Versions recorded
25
Latest version
v0.4, committed Oct 25, 2026
First version recorded
Sep 7, 2026
Size
56,356 bytes
Repository stars
33
License
Apache-2.0

Quantize any LLM the RIGHT way with Pollard Weights and ship it on every lane: GGUF (full ladder + trellis flagship), MLX, GPTQ, MX (NVFP4 / MXFP4 / FP8 / MXFP6 + mixes), EXL3, EXLR8 and DRAFT (speculative drafters), one repo per model. Use this whenever asked to quantize / shrink / "make a Pollard build" / "pollard it" / release a model, or to plan its fit on a machine or wafer. It routes dense vs MoE models down different paths — using the wrong one wastes hours.

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

# Pollard Weights — agent skill

Pollard makes a frontier model fit a given machine's memory by **deciding where the
bits go** instead of painting every weight the same. This skill tells an agent the
correct path for *this* model so results are good and no time is wasted.

## ⚠️ STEP 0 — ASK THE USER (do this before building; don't assume)

The machine you BUILD on is usually **not** the machine the model will RUN on (users build on a big-disk
box for a phone, a Pi, another server, a 5090, a wafer…). `--ram`/`--vram` is the **TARGET device's**
budget, NOT this machine's — so you cannot infer it. **Before building, ask the user two things** (unless
they already gave them in the prompt):

1. **Target device & budget** — where will the model RUN, and how much RAM/VRAM does it have? (e.g. "iPhone
   ~6GB", "Jetson 8GB", "a box with 24GB", "a 256GB server", "512GB wafer node"). Pollard targets **ANY**
   size — MB to TB — there is NO fixed 16–24GB range; that's just a common demo number.
2. **Preference** — *fit-a-budget* (best quality that fits that device, default) or **max-shrink** (smallest
   possible, quality be damned — `--ram <small>` / `--allow-1bit`)?
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add westwaters/pollard-weights/pollard-weights@sha256:38daa87716ffa4bc

The line at the top of this page pins the label v0.4, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_buac6gdshiweoscv.svg)](https://markdownregistry.com/a/art_buac6gdshiweoscv)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 25 versions of this file, between Sep 7, 2026 and Oct 25, 2026, about one every 2 days.

versioncommittedcommitsizeaudit
v0.4 latest2026-10-25 6222c54 56,356 BA view · diff
v0.42026-10-25 5b9feac 55,648 BA view · diff
v0.32026-10-16 45257b3 49,629 BA view · diff
v0.32026-10-15 16933f7 48,954 BA view · diff
v0.32026-10-15 4ab69f9 47,444 BA view · diff
v0.32026-10-15 98cf984 45,320 BA view · diff
v0.32026-10-15 9dbc402 45,182 BA view · diff
v0.32026-10-15 e7103bc 44,315 BA view · diff
v0.32026-10-15 e01afff 43,660 BA view · diff
v0.32026-10-15 7f712a1 42,787 BA view · diff
v0.32026-10-15 cf60ae4 40,720 BA view · diff
v0.32026-10-12 508c19e 36,722 BA view · diff
v0.32026-10-12 8709e9d 36,526 BA view · diff
v0.32026-10-10 beeea22 35,620 BA view · diff
v0.32026-10-05 67a9c61 35,506 BA view · diff
v0.32026-10-05 b679005 33,919 BA view · diff
v0.32026-09-16 7be030e 32,082 BA view · diff
v0.32026-09-16 5eca70c 32,081 BA view · diff
v0.32026-09-13 0978a40 28,685 BA view · diff
v0.32026-09-13 7726e1c 26,544 BA view · diff
v0.32026-09-07 d6730c1 23,527 BA view · diff
v0.32026-09-07 9b14269 23,010 BA view · diff
v0.32026-09-07 4bf948c 20,354 BA view · diff
v0.32026-09-07 72124dc 20,907 BA view · diff
v0.32026-09-07 78abb61 22,886 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (56356 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

GitHub

westwaters/pollard-weights · 33 stars · license Apache-2.0 · pushed 2026-11-04 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_buac6gdshiweoscv
GET https://markdownregistry.com/api/v1/resolve?ref=westwaters/pollard-weights/pollard-weights
GET https://markdownregistry.com/api/v1/blob/38daa87716ffa4bcf460faff00d794d0690be57b0dee863de3a95fe3bca90d0e

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.

More from westwaters/pollard-weights

fine-tune-and-serve SKILL.md skill
westwaters/pollard-weights · skills/pollard/fine-tune-and-serve/SKILL.md · Fine-tune an open model (Qwen, Llama, ...) on the user's OWN data with Pollard, prove on a held-out set that it beats…
v0.2 · audit A · 33 stars

Every file in westwaters/pollard-weights

Browse by kind, by grade A, or by owner.