# pytorch/torchchat

Run PyTorch LLMs locally on servers, desktop and mobile

Repository: https://github.com/pytorch/torchchat
Canonical: https://ross.abutalabs.com/products/torchchat
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: llm, local, pytorch
Archived: true
Last push: 2025-09-10T22:22:42+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 41, release rhythm 35, longevity 63
- inputs: {"age_days": 894, "days_push": 357, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3612, forks 248 (observed 2026-08-28T04:08:11.675202+00:00)

## What it is
torchchat is a compact PyTorch-native codebase for running large language models locally via Python, a C/C++ runner on desktop/server, or on iOS and Android devices. It supports popular models like Llama 3, Mistral, and DeepSeek R1 Distill with multiple quantization and execution modes.

## Use cases
- run llama 3 locally on my mac
- chat with an llm from the command line
- deploy a small language model on android or ios
- run llm inference in c++ without python
- quantize and evaluate llms locally with pytorch
- run multimodal llama 3.2 vision model locally

## When to choose
- you want PyTorch-native local LLM inference across server, desktop, and mobile
- you need to run LLMs in C/C++ applications without a Python dependency
- you want on-device LLM deployment on iOS or Android via ExecuTorch

## When to avoid
- you need actively developed tooling - the project is no longer under active development
- you want GPU-cluster-scale inference or serving many concurrent users
- you need broad model support beyond the Llama-family and a few supported models

## Facets
- artifact type: cli-tool
- maturity: maintenance
- function: llm-inference, cli, chatbot
- domain: large-language-models, machine-learning, developer-tools, mobile-development
- platform: python, cli, cpp
- tags: pytorch, local-inference, llm, on-device-ai, aot-inductor, executorch, quantization, multimodal, linux, macos, android, ios

## Member repositories
- pytorch/torchchat (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:11.675202+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 2026-08-29T18:33:38.628618+00:00, confidence not recorded.
  - readme: https://github.com/pytorch/torchchat (fetched 2026-08-28T04:08:11.675202+00:00, sha cf1310bb39fd)
- Data as of 2026-08-30T08:39:29.467469+00:00.