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

PalateKit

Menu

Request a resource

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

0 / 280

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.

Launch Megatron-LM on a SLURM cluster v1.0.0

Use this resource to launch distributed Megatron-LM training on a SLURM cluster and monitor its ranks

New

  • 600 downloads
  • 239 likes
  • 101 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install megatron-distributed-slurm-launch

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 writes an sbatch script matched to your node and GPU count, sets the launch variables and picks the CUDA_DEVICE_MAX_CONNECTIONS setting according to the hardware and parallelism mode. It also helps diagnose a failing rank.

Mostly picked by

Updated November 2, 2026

Card for the “Launch Megatron-LM on a SLURM cluster” skill

Details

Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md : sbatch script, settings and diagnosis
  • evals/evals.json : evaluation cases for the skill

Good fit for

  • You scale a Megatron-LM training from one node to several
  • A rank fails on a cluster and you need the origin

Not for

    Tags

    • megatron-lm
    • slurm
    • sbatch
    • distributed
    • gpu

    About

    You say

    “Write the sbatch script to launch Megatron-LM on four nodes with eight GPUs each.”

    Card for the “Launch Megatron-LM on a SLURM cluster” 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 launch distributed Megatron-LM training on a SLURM cluster and monitor its ranks
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md2.8 KB
    • evals/evals.json0 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • skill-card.md2.9 KB
    • SKILL.md7.2 KB
    • skill.oms.sig4.5 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/mcore-run-on-slurm
    Licence
    Apache-2.0

    Notice

    « mcore-run-on-slurm » (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.