# mistralai/mistral-finetune

Repository: https://github.com/mistralai/mistral-finetune
Canonical: https://ross.abutalabs.com/products/mistral-finetune
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2026-06-16T15:22:10+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 87, release rhythm 35, longevity 59
- inputs: {"age_days": 831, "days_push": 78, "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 3096, forks 317 (observed 2026-08-28T04:07:43.387913+00:00)

## What it is
A lightweight Python codebase from Mistral AI for memory-efficient LoRA fine-tuning of Mistral's language models, optimized for single-node multi-GPU setups. The repository is now archived and no longer maintained.

## Use cases
- fine-tune Mistral 7B on my own data
- LoRA fine-tuning of Mistral models
- train a custom chat model from a Mistral checkpoint
- memory-efficient LLM fine-tuning on a single A100
- adapt Mistral Large or Nemo to my domain
- instruction-tune a Mistral instruct model

## When to choose
- You specifically want to fine-tune Mistral models with LoRA and have A100/H100 GPUs
- You want a simple, opinionated entrypoint for Mistral fine-tuning with guided data formatting

## When to avoid
- You need actively maintained software - the repo is archived
- You need support for many model architectures or hardware types
- You want a generic fine-tuning framework - consider torchtune instead

## Facets
- artifact type: library
- maturity: abandoned
- function: llm-training, machine-learning, deep-learning
- domain: large-language-models, machine-learning, deep-learning
- platform: python
- tags: lora, fine-tuning, mistral, archived, pytorch, gpu, linux

## Member repositories
- mistralai/mistral-finetune (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:43.387913+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 2026-08-29T18:46:23.217578+00:00, confidence not recorded.
  - readme: https://github.com/mistralai/mistral-finetune (fetched 2026-08-28T04:07:43.387913+00:00, sha df2164b1f85b)
- Data as of 2026-08-30T08:39:29.467469+00:00.