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.

Distributed training with DeepSpeed v1.0.0

Use this resource to set up distributed training with DeepSpeed

New

  • 309 downloads
  • 229 likes
  • 232 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install deepspeed-distributed-training-guide

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 helps you choose a ZeRO stage, a precision and a parallelism mode for DeepSpeed, then writes the matching code snippets. You get a configuration and examples drawn from the reference documentation.

Mostly picked by

Updated November 2, 2026

Card for the “Distributed training with DeepSpeed” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: use cases, quick reference and common code patterns
  • references/: documentation notes by topic, including DeepNVMe
  • references/: tutorials and resources from the DeepSpeed documentation

Good fit for

  • Your model does not fit in the memory of a single GPU
  • You want to reduce optimizer memory with ZeRO

Not for

    Tags

    • deepspeed
    • zero
    • distributed training
    • bf16

    About

    You say

    “Set up DeepSpeed ZeRO stage 3 in BF16 to train my 13-billion-parameter model on 4 GPUs.”

    Card for the “Distributed training with DeepSpeed” skill

    Skill published by Orchestra Research under the MIT 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 set up distributed training with DeepSpeed
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/08.md0.3 KB
    • references/09.md27 KB
    • references/2020.md35.3 KB
    • references/2023.md11.8 KB
    • references/assets.md2.6 KB
    • references/index.md0.4 KB
    • references/mii.md10.9 KB
    • references/other.md98.5 KB
    • references/tutorials.md450.2 KB
    • SKILL.md141.7 KB

    Origin and licence

    Listed unmodified

    Author
    Orchestra Research
    Source
    github.com/davila7/claude-code-templates/tree/f57964ebda91cdfb31755661ecea9395f30020f1/cli-tool/components/skills/ai-research/distributed-training-deepspeed
    Licence
    MIT

    Notice

    « distributed-training-deepspeed » (Orchestra Research, MIT) 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.