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.

Domain parallelism with PhysicsNeMo ShardTensor v1.0.0

Use this resource to integrate ShardTensor domain parallelism into a PhysicsNeMo training

New

  • 499 downloads
  • 178 likes
  • 115 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install domain-parallel-shard-tensor

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 sets up the GPU mesh, shards the input along the spatial dimension, registers the patches needed by new layers and bootstraps multi-GPU tests. It notes that neither DDP nor FSDP2 syncs weights across the domain axis.

Mostly picked by

Updated November 2, 2026

Card for the “Domain parallelism with PhysicsNeMo ShardTensor” 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: integration principles and procedure
  • references/integration-checklist.md: integration checklist
  • references/testing.md: multi-GPU tests

Good fit for

  • Your model does not fit on one GPU because of image or sequence size
  • You add a new operation to a sharded model

Not for

    Tags

    • shardtensor
    • parallelism
    • physicsnemo
    • multi-gpu

    About

    You say

    “Shard my BCHW input along the height and integrate ShardTensor into my DDP script.”

    Card for the “Domain parallelism with PhysicsNeMo ShardTensor” 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 integrate ShardTensor domain parallelism into a PhysicsNeMo training
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md4.6 KB
    • evals/evals.json6.5 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/integration-checklist.md4.1 KB
    • references/new-op-patterns.md7.4 KB
    • references/testing.md9.6 KB
    • skill-card.md4.3 KB
    • SKILL.md13.8 KB
    • skill.oms.sig5.2 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/physicsnemo-shard-tensor
    Licence
    Apache-2.0

    Notice

    « physicsnemo-shard-tensor » (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.