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.

Training Mixture of Experts models v1.0.0

Use this resource to train a sparse experts model with routing and load balancing

New

  • 499 downloads
  • 228 likes
  • 133 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install moe-mixture-of-experts-training

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 guides you through configuring an MoE model with DeepSpeed or Hugging Face, defining routing and load balancing, then launching training. It also covers architectures such as Mixtral and inference optimization.

Mostly picked by

Updated November 2, 2026

Card for the “Training Mixture of Experts models” skill

Details

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

Similar resources

View all

What you get

  • SKILL.md: installation, basic MoE architecture and training script
  • references/architectures.md, training.md and inference.md: architectures, training and inference

Good fit for

  • You want more capacity without a proportional increase in compute
  • You study architectures such as Mixtral or DeepSeek-V3

Not for

    Tags

    • moe
    • experts
    • routing
    • distributed

    About

    You say

    “Prepare a four-expert MoE training run with DeepSpeed and load balancing.”

    Card for the “Training Mixture of Experts models” 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 train a sparse experts model with routing and load balancing
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/architectures.md14.1 KB
    • references/inference.md8.4 KB
    • references/training.md9.9 KB
    • SKILL.md15.1 KB

    Origin and licence

    Listed unmodified

    Author
    Orchestra Research
    Source
    github.com/orchestra-research/ai-research-skills/tree/773a52944ba4747a18bd4ae9ade53fff041adcbc/19-emerging-techniques/moe-training
    Licence
    MIT

    Notice

    « moe-training » (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.