distributed-llm-pretraining-torchtitan
Run large-scale LLM pretraining using PyTorch's official torchtitan platform with 4D parallelism.
Install
mkdir -p .claude/skills/distributed-llm-pretraining-torchtitan && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/6227" && unzip -o skill.zip -d .claude/skills/distributed-llm-pretraining-torchtitan && rm skill.zipInstalls to .claude/skills/distributed-llm-pretraining-torchtitan
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.Key capabilities
- →Pretrain LLMs using 4D parallelism
- →Apply Float8 quantization for H100 GPUs
- →Manage distributed checkpointing and resharding
- →Configure torch.compile for training optimization
- →Scale model training across 8 to 512+ GPUs
How it works
It utilizes PyTorch-native composable parallelism including FSDP2, Tensor, Pipeline, and Context Parallelism to distribute model training across multiple GPUs. It supports Float8 precision and compilation to accelerate large-scale pretraining.
Inputs & outputs
When to use distributed-llm-pretraining-torchtitan
- →Pretrain Llama 3.1 8B models
- →Scale training across 512+ GPUs
- →Apply Float8 quantization to training runs
- →Configure 4D model parallelism
About this skill
TorchTitan - PyTorch Native Distributed LLM Pretraining
Quick start
TorchTitan is PyTorch's official platform for large-scale LLM pretraining with composable 4D parallelism (FSDP2, TP, PP, CP), achieving 65%+ speedups over baselines on H100 GPUs.
Installation:
# From PyPI (stable)
pip install torchtitan
# From source (latest features, requires PyTorch nightly)
git clone https://github.com/pytorch/torchtitan
cd torchtitan
pip install -r requirements.txt
Download tokenizer:
# Get HF token from https://huggingface.co/settings/tokens
python scripts/download_hf_assets.py --repo_id meta-llama/Llama-3.1-8B --assets tokenizer --hf_token=...
Start training on 8 GPUs:
CONFIG_FILE="./torchtitan/models/llama3/train_configs/llama3_8b.toml" ./run_train.sh
Common workflows
Workflow 1: Pretrain Llama 3.1 8B on single node
Copy this checklist:
Single Node Pretraining:
- [ ] Step 1: Download tokenizer
- [ ] Step 2: Configure training
- [ ] Step 3: Launch training
- [ ] Step 4: Monitor and checkpoint
Step 1: Download tokenizer
python scripts/download_hf_assets.py \
--repo_id meta-llama/Llama-3.1-8B \
--assets tokenizer \
--hf_token=YOUR_HF_TOKEN
Step 2: Configure training
Edit or create a TOML config file:
# llama3_8b_custom.toml
[job]
dump_folder = "./outputs"
description = "Llama 3.1 8B training"
[model]
name = "llama3"
flavor = "8B"
hf_assets_path = "./assets/hf/Llama-3.1-8B"
[optimizer]
name = "AdamW"
lr = 3e-4
[lr_scheduler]
warmup_steps = 200
[training]
local_batch_size = 2
seq_len = 8192
max_norm = 1.0
steps = 1000
dataset = "c4"
[parallelism]
data_parallel_shard_degree = -1 # Use all GPUs for FSDP
[activation_checkpoint]
mode = "selective"
selective_ac_option = "op"
[checkpoint]
enable = true
folder = "checkpoint"
interval = 500
Step 3: Launch training
# 8 GPUs on single node
CONFIG_FILE="./llama3_8b_custom.toml" ./run_train.sh
# Or explicitly with torchrun
torchrun --nproc_per_node=8 \
-m torchtitan.train \
--job.config_file ./llama3_8b_custom.toml
Step 4: Monitor and checkpoint
TensorBoard logs are saved to ./outputs/tb/:
tensorboard --logdir ./outputs/tb
Workflow 2: Multi-node training with SLURM
Multi-Node Training:
- [ ] Step 1: Configure parallelism for scale
- [ ] Step 2: Set up SLURM script
- [ ] Step 3: Submit job
- [ ] Step 4: Resume from checkpoint
Step 1: Configure parallelism for scale
For 70B model on 256 GPUs (32 nodes):
[parallelism]
data_parallel_shard_degree = 32 # FSDP across 32 ranks
tensor_parallel_degree = 8 # TP within node
pipeline_parallel_degree = 1 # No PP for 70B
context_parallel_degree = 1 # Increase for long sequences
Step 2: Set up SLURM script
#!/bin/bash
#SBATCH --job-name=llama70b
#SBATCH --nodes=32
#SBATCH --ntasks-per-node=8
#SBATCH --gpus-per-node=8
srun torchrun \
--nnodes=32 \
--nproc_per_node=8 \
--rdzv_backend=c10d \
--rdzv_endpoint=$MASTER_ADDR:$MASTER_PORT \
-m torchtitan.train \
--job.config_file ./llama3_70b.toml
Step 3: Submit job
sbatch multinode_trainer.slurm
Step 4: Resume from checkpoint
Training auto-resumes if checkpoint exists in configured folder.
Workflow 3: Enable Float8 training for H100s
Float8 provides 30-50% speedup on H100 GPUs.
Float8 Training:
- [ ] Step 1: Install torchao
- [ ] Step 2: Configure Float8
- [ ] Step 3: Launch with compile
Step 1: Install torchao
USE_CPP=0 pip install git+https://github.com/pytorch/ao.git
Step 2: Configure Float8
Add to your TOML config:
[model]
converters = ["quantize.linear.float8"]
[quantize.linear.float8]
enable_fsdp_float8_all_gather = true
precompute_float8_dynamic_scale_for_fsdp = true
filter_fqns = ["output"] # Exclude output layer
[compile]
enable = true
components = ["model", "loss"]
Step 3: Launch with compile
CONFIG_FILE="./llama3_8b.toml" ./run_train.sh \
--model.converters="quantize.linear.float8" \
--quantize.linear.float8.enable_fsdp_float8_all_gather \
--compile.enable
Workflow 4: 4D parallelism for 405B models
4D Parallelism (FSDP + TP + PP + CP):
- [ ] Step 1: Create seed checkpoint
- [ ] Step 2: Configure 4D parallelism
- [ ] Step 3: Launch on 512 GPUs
Step 1: Create seed checkpoint
Required for consistent initialization across PP stages:
NGPU=1 CONFIG_FILE=./llama3_405b.toml ./run_train.sh \
--checkpoint.enable \
--checkpoint.create_seed_checkpoint \
--parallelism.data_parallel_shard_degree 1 \
--parallelism.tensor_parallel_degree 1 \
--parallelism.pipeline_parallel_degree 1
Step 2: Configure 4D parallelism
[parallelism]
data_parallel_shard_degree = 8 # FSDP
tensor_parallel_degree = 8 # TP within node
pipeline_parallel_degree = 8 # PP across nodes
context_parallel_degree = 1 # CP for long sequences
[training]
local_batch_size = 32
seq_len = 8192
Step 3: Launch on 512 GPUs
# 64 nodes x 8 GPUs = 512 GPUs
srun torchrun --nnodes=64 --nproc_per_node=8 \
-m torchtitan.train \
--job.config_file ./llama3_405b.toml
When to use vs alternatives
Use TorchTitan when:
- Pretraining LLMs from scratch (8B to 405B+)
- Need PyTorch-native solution without third-party dependencies
- Require composable 4D parallelism (FSDP2, TP, PP, CP)
- Training on H100s with Float8 support
- Want interoperable checkpoints with torchtune/HuggingFace
Use alternatives instead:
- Megatron-LM: Maximum performance for NVIDIA-only deployments
- DeepSpeed: Broader ZeRO optimization ecosystem, inference support
- Axolotl/TRL: Fine-tuning rather than pretraining
- LitGPT: Educational, smaller-scale training
Common issues
Issue: Out of memory on large models
Enable activation checkpointing and reduce batch size:
[activation_checkpoint]
mode = "full" # Instead of "selective"
[training]
local_batch_size = 1
Or use gradient accumulation:
[training]
local_batch_size = 1
global_batch_size = 32 # Accumulates gradients
Issue: TP causes high memory with async collectives
Set environment variable:
export TORCH_NCCL_AVOID_RECORD_STREAMS=1
Issue: Float8 training not faster
Float8 only benefits large GEMMs. Filter small layers:
[quantize.linear.float8]
filter_fqns = ["attention.wk", "attention.wv", "output", "auto_filter_small_kn"]
Issue: Checkpoint loading fails after parallelism change
Use DCP's resharding capability:
# Convert sharded checkpoint to single file
python -m torch.distributed.checkpoint.format_utils \
dcp_to_torch checkpoint/step-1000 checkpoint.pt
Issue: Pipeline parallelism initialization
Create seed checkpoint first (see Workflow 4, Step 1).
Supported models
| Model | Sizes | Status |
|---|---|---|
| Llama 3.1 | 8B, 70B, 405B | Production |
| Llama 4 | Various | Experimental |
| DeepSeek V3 | 16B, 236B, 671B (MoE) | Experimental |
| GPT-OSS | 20B, 120B (MoE) | Experimental |
| Qwen 3 | Various | Experimental |
| Flux | Diffusion | Experimental |
Performance benchmarks (H100)
| Model | GPUs | Parallelism | TPS/GPU | Techniques |
|---|---|---|---|---|
| Llama 8B | 8 | FSDP | 5,762 | Baseline |
| Llama 8B | 8 | FSDP+compile+FP8 | 8,532 | +48% |
| Llama 70B | 256 | FSDP+TP+AsyncTP | 876 | 2D parallel |
| Llama 405B | 512 | FSDP+TP+PP | 128 | 3D parallel |
Advanced topics
FSDP2 configuration: See references/fsdp.md for detailed FSDP2 vs FSDP1 comparison and ZeRO equivalents.
Float8 training: See references/float8.md for tensorwise vs rowwise scaling recipes.
Checkpointing: See references/checkpoint.md for HuggingFace conversion and async checkpointing.
Adding custom models: See references/custom-models.md for TrainSpec protocol.
Resources
- GitHub: https://github.com/pytorch/torchtitan
- Paper: https://arxiv.org/abs/2410.06511
- ICLR 2025: https://iclr.cc/virtual/2025/poster/29620
- PyTorch Forum: https://discuss.pytorch.org/c/distributed/torchtitan/44
When not to use it
- →Fine-tuning tasks
- →Small-scale training projects
- →Scenarios requiring third-party dependencies like Megatron-LM
Prerequisites
Limitations
- →Requires specific hardware for Float8 benefits
- →Checkpoint loading may fail if parallelism settings change without resharding
How it compares
This workflow provides a native PyTorch platform for pretraining that avoids the complexity of third-party distributed training frameworks.
Compared to similar skills
distributed-llm-pretraining-torchtitan side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| distributed-llm-pretraining-torchtitan (this skill) | 1 | 6mo | Review | Advanced |
| unsloth | 15 | 8mo | No flags | Intermediate |
| llm-application-dev | 3 | 4mo | Review | Intermediate |
| book-sft-pipeline | 3 | 3mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by davila7
View all by davila7 →You might also like
unsloth
zechenzhangAGI
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
llm-application-dev
skillcreatorai
Building applications with Large Language Models - prompt engineering, RAG patterns, and LLM integration. Use for AI-powered features, chatbots, or LLM-based automation.
book-sft-pipeline
muratcankoylan
This skill should be used when the user asks to "fine-tune on books", "create SFT dataset", "train style model", "extract ePub text", or mentions style transfer, LoRA training, book segmentation, or author voice replication.
fine-tuning-with-trl
davila7
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
huggingface-tokenizers
davila7
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
massgen-develops-massgen
massgen
Guide for using MassGen to develop and improve itself. This skill should be used when agents need to run MassGen experiments programmatically (using automation mode) OR analyze terminal UI/UX quality (using visual evaluation tools). These are mutually exclusive workflows for different improvement goals.