distributed-llm-pretraining-torchtitan

distributed-llm-pretraining-torchtitan is a skill for Claude Code, Codex from math-inc/OpenGauss. It costs 83 tokens per session (2,641 once invoked), scanned A, a copy of distributed-llm-pretraining-torchtitan, MIT.

A PyTorch platform for training large language models across many GPUs at once. It divides the model, data, and sequences across the available hardware.

In plain words
What is it for?
Use it to pretrain Llama, DeepSeek, or custom language models, manage distributed checkpoints, and configure training across 8 to 512 or more GPUs.
Why use it?
It helps run pretraining jobs that are too large for one GPU or one computer.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python scripts/download_hf_assets.py --repo_id meta-llama/Llama-3.1-8B --assets tokenizer --hf_token=....

Good fit Use it to pretrain Llama, DeepSeek, or custom language models, manage distributed checkpoints, and configure training across 8 to 512 or more GPUs.

Compare 6 skills from other repositories ↓
About the project

OpenGauss is a project-scoped Lean workflow orchestrator that gives coding agents a command-line interface for managing formal proof and formalization tasks. It is used with Lean projects to coordinate agents, tooling, backend sessions, and workflows supplied by lean4-skills. The catalogue add-ons operate these Gauss-native workflows.

math-inc/OpenGauss · 1,260 stars · on GitHub

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/math-inc/OpenGauss
agentmods
npx agentmods add skills/math-inc/opengauss/torchtitan

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for distributed-llm-pretraining-torchtitan

README.md
[![agentmods](https://agentmods.dev/badge/skills/math-inc/opengauss/torchtitan/github.svg)](https://agentmods.dev/skills/math-inc/opengauss/torchtitan)
Your own site
<a href="https://agentmods.dev/skills/math-inc/opengauss/torchtitan"><img src="https://agentmods.dev/badge/skills/math-inc/opengauss/torchtitan/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for distributed-llm-pretraining-torchtitan

Your own site · 80×15
<a href="https://agentmods.dev/skills/math-inc/opengauss/torchtitan"><img src="https://agentmods.dev/badge/skills/math-inc/opengauss/torchtitan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 83 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,641 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 94% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00083 $0.02641
Opus 5 $0.00042 $0.01321
Sonnet 5 $0.00017 $0.00528
Haiku 4.5 $0.00008 $0.00264

Measured 6d ago against content hash 79c2cfbbe99c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

distributed-llm-pretraining-torchtitan scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

94% identical to distributed-llm-pretraining-torchtitan — 5 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/mlops/training/torchtitan/SKILL.md · 362 lines

How it starts

The opening of the file, as written. The whole thing — 362 lines — stays where its author put it; the contents beside it link to each section on GitHub.

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

Read the full file on GitHub · 362 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 362 lines · 83 tokens per session scan A 79c2cfbbe99c

Subscribe to this mod's changes

distributed-llm-pretraining-torchtitan is a skill published in the GitHub repository math-inc/OpenGauss (1,260 stars, last pushed 5mo ago), licensed MIT. It adds 83 tokens to every session and 2,641 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to distributed-llm-pretraining-torchtitan, differing in 5 lines, and is treated as a copy.

Related

Other skills, from other repositories

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

agent-platform-model-registry

Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.

google/skills · 60 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens