claudegoodies
Skill

moe-training

From davila7

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Facts

Repository
davila7/claude-code-templates
Status
Actively maintained
Last commit
Source file
cli-tool/components/skills/ai-research/emerging-techniques-moe-training/SKILL.md

Source preview

The instructions Claude Code reads when this skill runs.

# MoE Training: Mixture of Experts

## When to Use This Skill

Use MoE Training when you need to:
- **Train larger models** with limited compute (5× cost reduction vs dense models)
- **Scale model capacity** without proportional compute increase
- **Achieve better performance** per compute budget than dense models
- **Specialize experts** for different domains/tasks/languages
- **Reduce inference latency** with sparse activation (only 13B/47B params active in Mixtral)
- **Implement SOTA models** like Mixtral 8x7B, DeepSeek-V3, Switch Transformers

**Notable MoE Models**: Mixtral 8x7B (Mistral AI), DeepSeek-V3, Switch Transformers (Google), GLaM (Google), NLLB-MoE (Meta)

## Installation

```bash
# DeepSpeed with MoE support
pip install deepspeed>=0.6.0

# Megatron-DeepSpeed for large-scale training
git clone https://github.com/microsoft/Megatron-DeepSpeed
cd Megatron-DeepSpeed
pip install -r requirements.txt

# Alternative: HuggingFace Transformers
pip install transformers accelerate
```

## Quick Start

### Basic MoE Architecture

```python
import torch
import torch.nn as nn

class MoELayer(nn.Module):
    """Sparse Mixture of Experts layer."""

    def __init__(self, hidden_size, num_experts=8, top_k=2):
        super().__init__()
        self.num_experts = num_experts
        self.top_k = top_k

        # Expert networks (FFN)
        self.experts = nn.ModuleList([
View full source on GitHub →

Reused elsewhere

An identical copy of this skill ships in 2 repositories.

Other skills