# pytorch/torchtitan

A PyTorch native platform for training generative AI models

Repository: https://github.com/pytorch/torchtitan
Canonical: https://ross.abutalabs.com/products/torchtitan
Language: Python
License: BSD-3-Clause
License Family: permissive
Last push: 2026-08-26T23:25:56+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 59, longevity 71
- inputs: {"age_days": 995, "days_push": 7, "days_rel": 194, "gap_med": 69, "n_releases_24m": 4}
- flags: prerelease_only
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5667, forks 968 (observed 2026-08-28T04:09:27.580142+00:00)

## What it is
torchtitan is a PyTorch-native platform for large-scale training of generative AI models, offering a clean-room implementation of PyTorch's distributed scaling techniques such as multi-dimensional parallelism. It supports pretraining LLMs like Llama 3.1 and includes an experimental RL training stack (TitanRL) that integrates with vLLM.

## Use cases
- pretrain large language models like Llama 3.1 on GPU clusters
- experiment with multi-dimensional parallelism in PyTorch
- train generative AI models at scale with minimal code changes
- run reinforcement learning training with vLLM generation
- benchmark PyTorch distributed training features
- extend a minimal training codebase with custom model architectures

## When to choose
- you want a clean, hackable PyTorch-native codebase for LLM pretraining
- you need to apply FSDP, tensor parallelism, or pipeline parallelism with minimal model code changes
- you are researching new model architectures or distributed training techniques
- you want to stay on the latest PyTorch features and nightlies

## When to avoid
- you need a turnkey fine-tuning product with a high-level API like Axolotl or LLaMA-Factory
- you require stable long-term support on older PyTorch releases
- you are not training on GPUs or lack multi-GPU infrastructure
- you need production serving or inference rather than training

## Facets
- artifact type: framework
- maturity: active
- function: llm-training, machine-learning, deep-learning, gpu-computing, benchmarking
- domain: large-language-models, deep-learning, machine-learning, gpu-computing, developer-tools
- platform: python, cloud
- tags: pytorch, distributed-training, llm-pretraining, parallelism, reinforcement-learning, generative-ai, gpu, linux, docker

## Member repositories
- pytorch/torchtitan (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:27.580142+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:54:18.786675+00:00, confidence not recorded.
  - readme: https://github.com/pytorch/torchtitan (fetched 2026-08-28T04:09:27.580142+00:00, sha f51cf888b2ee)
  - registry_pypi: https://pypi.org/pypi/torchtitan/json (fetched 2026-08-29T08:49:30.746549+00:00, sha 761f777cf7ff)
- Data as of 2026-08-30T08:39:29.467469+00:00.