# mlabonne/llm-datasets

Curated list of datasets and tools for post-training.

Repository: https://github.com/mlabonne/llm-datasets
Canonical: https://ross.abutalabs.com/products/llm-datasets
Homepage: https://mlabonne.github.io/blog
License Family: other
Topics: data, dataset, llm
Last push: 2026-04-29T18:17:40+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 79, release rhythm 35, longevity 61
- inputs: {"age_days": 858, "days_push": 126, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4757, forks 394 (observed 2026-08-28T04:08:58.839329+00:00)

## What it is
A curated list of datasets and tools for post-training large language models, covering supervised fine-tuning, preference, and reasoning datasets. It catalogs permissively licensed instruction datasets with metadata like sample counts, thinking traces, and licensing notes.

## Use cases
- find datasets for fine-tuning an LLM
- curated list of instruction tuning datasets
- datasets for supervised fine-tuning of language models
- post-training data for LLMs
- where to get SFT training data
- reasoning datasets with verified traces
- compare open LLM training datasets

## When to choose
- you are selecting training data for fine-tuning or post-training an LLM
- you want a regularly updated catalog of permissively licensed instruction datasets
- you need guidance on what makes a high-quality LLM dataset

## When to avoid
- you need a tool that generates or processes datasets rather than a list of them
- you want pre-trained models or inference tooling instead of training data
- you need a formal benchmark rather than a curated catalog

## Facets
- artifact type: learning-resource
- maturity: active
- function: llm-training, data-science, etl
- domain: large-language-models, machine-learning, artificial-intelligence, awesome-lists, tutorials
- platform: cross-platform
- tags: curated-list, datasets, post-training, fine-tuning, sft, instruction-datasets, llm

## Member repositories
- mlabonne/llm-datasets (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:58.839329+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 2026-08-29T18:18:52.717164+00:00, confidence not recorded.
  - readme: https://github.com/mlabonne/llm-datasets (fetched 2026-08-28T04:08:58.839329+00:00, sha 31a176059a5d)
  - homepage: https://mlabonne.github.io/blog (fetched 2026-08-29T09:02:41.261504+00:00, sha 4fd183164a78)
- Data as of 2026-08-30T08:39:29.467469+00:00.