gpu-dispatch SKILL.md skillA
gpu-dispatch from tcuzzo/backs-aios-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add tcuzzo/backs-aios-skills/gpu-dispatch@git:20260828.045e846Pins git:20260828.045e846 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- tcuzzo/backs-aios-skills
- Path
i18n/de/skills/gpu-dispatch/SKILL.md- Versions recorded
- 3
- Latest version
- git:20260828.045e846, committed Aug 28, 2026
- First version recorded
- Aug 27, 2026
- Size
- 4,738 bytes
- Repository stars
- 2
- License
- a license file GitHub does not identify
Nutze das beim Dispatch lokaler Modelle auf GPUs — Inferenzarbeit planen, eine Karte wählen oder Modell-Residenz verwalten. Ein Modell pro GPU, kein Spill ins System-RAM, warm halten durch die Schleife, am Schleifenende entladen, Zulassung nach gemessener Wahrheit. Trigger words: gpu, vram, gpu dispatch, model loading, keep alive, resident model, local inference, spill, warm, Grafikkarte, Modell laden, lokale Inferenz, warmhalten.
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
# Das GPU-Dispatch-Gesetz **Effort:** free — vom Dispatcher erzwungene Regeln, abgelesen am eigenen Live-Zustand des Knotens; senkt die Kosten unterm Strich. Beseitigt: Läufe, die mit übergelaufenem VRAM still 10x langsamer sind, Kaltstart-Wechselei zwischen Jobs und Karten, die wegen vermuteter Zäune brachliegen. Vier Regeln für lokale Modelle auf GPUs. Es gibt sie, weil die zwei häufigen Fehlermodi entgegengesetzt und gleich teuer sind: Karten mit Loads und Spills zu schreddern, und Hardware so einzuzäunen, dass sie brachliegt. Beides ist verlorene Fähigkeit. Erzwinge diese Regeln im Dispatcher als Code — nie als Regel, die sich ein Modell merken muss. ## Wann einsetzen - Vor dem Dispatch jedes Inferenzjobs auf eine lokale GPU. - Beim Entwerfen oder Reviewen eines Dispatchers, Schedulers oder Modell-Routers. - Wenn ein lokaler Lauf rätselhaft langsam ist, oder eine Karte rätselhaft „nicht verfügbar". ## Die vier Regeln 1. **Ein Modell resident pro Karte, zur selben Zeit.** Lies vor jedem Dispatch den Live-Zustand geladener Modelle des Knotens aus der eigenen API der Runtime. Ist ein anderes Modell resident, nutze es oder entlade es zuerst. Lade nie ein …
Read the whole file at its exact version.
How to install
mdr add tcuzzo/backs-aios-skills/gpu-dispatch@sha256:440f3e0e272391dcThe line at the top of this page pins the label git:20260828.045e846, 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_x4v5g654nevly2fm)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Aug 27, 2026 and Aug 28, 2026, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260828.045e846 latest | 2026-08-28 | 045e846 | 4,738 B | A | view · diff |
| git:20260827.a68a1ea | 2026-08-27 | a68a1ea | 4,732 B | A | view · diff |
| git:20260827.e618bb3 | 2026-08-27 | e618bb3 | 4,437 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 (4738 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
tcuzzo/backs-aios-skills · 2 stars · license a license file GitHub does not identify · pushed 2026-11-03 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_x4v5g654nevly2fm GET https://markdownregistry.com/api/v1/resolve?ref=tcuzzo/backs-aios-skills/gpu-dispatch GET https://markdownregistry.com/api/v1/blob/440f3e0e272391dca91902ffd520a3c8c08e6882549c010272432aeb21ff7ea4
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.