Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

CPU offloading in Megatron Bridge v1.0.0

Use this resource to enable and validate CPU offloading to reduce GPU memory

New

  • 355 downloads
  • 234 likes
  • 241 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install cpu-offloading-megatron-bridge

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent separates activation offloading, which requires PP equal to 1, from optimizer offloading, which has no PP constraint. It recommends the second mode for large models and fills the matching parameters. You get a configuration and its compatibility constraints.

Mostly picked by

Updated November 2, 2026

Card for the “CPU offloading in Megatron Bridge” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md: method and constraints
  • card.yaml: structured card
  • evals/evals.json: evaluation cases

Good fit for

  • A large model does not fit with optimizer states on the GPU
  • You want to offload activations of a medium-sized model

Not for

    Tags

    • CPU offloading
    • optimizer
    • memory
    • Megatron Bridge

    About

    You say

    “Enable optimizer offloading on a 70B model and tell me what it requires from parallelism.”

    Card for the “CPU offloading in Megatron Bridge” skill

    Skill published by nvidia under the Apache-2.0 license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to enable and validate CPU offloading to reduce GPU memory
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md4.3 KB
    • card.yaml8.3 KB
    • evals/evals.json1.5 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • skill-card.md3.9 KB
    • SKILL.md9.3 KB
    • skill.oms.sig4.7 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/nemo-mbridge-perf-cpu-offloading
    Licence
    Apache-2.0

    Notice

    « nemo-mbridge-perf-cpu-offloading » (nvidia, Apache-2.0) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.