# instructlab/instructlab

InstructLab Core package.  Use this to chat with a model and execute the InstructLab workflow to train a model using custom taxonomy data.

Repository: https://github.com/instructlab/instructlab
Canonical: https://ross.abutalabs.com/products/instructlab
Homepage: https://instructlab.ai
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2026-03-30T22:38:31+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 74, release rhythm 40, longevity 66
- inputs: {"age_days": 924, "days_push": 156, "days_rel": 485, "gap_med": 5, "n_releases_24m": 28}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1419, forks 455 (observed 2026-08-28T04:04:40.396477+00:00)

## What it is
InstructLab (ilab) is an open-source CLI and Python package for enhancing large language models through community-contributed taxonomy-based skills and knowledge. It uses the LAB alignment method to generate synthetic data from taxonomy contributions and fine-tune base models, supporting local chatting, serving, data generation, and training workflows.

## Use cases
- fine-tune an LLM with custom skills and knowledge
- generate synthetic training data from a taxonomy of examples
- chat with a locally served quantized model
- contribute new skills to a community LLM via pull requests
- run the LAB alignment tuning method on a base model
- serve and test a tuned model locally

## When to choose
- you want to augment an open LLM with community-contributed skills via synthetic data
- you need an end-to-end local workflow for taxonomy-based tuning, serving, and chatting
- you want to contribute model improvements without retraining from scratch

## When to avoid
- you need a fully managed or hosted fine-tuning service
- you only want inference without any model tuning
- you require the newest SDG or training components, which are moving to separate sdg_hub and training_hub repositories

## Facets
- artifact type: cli-tool
- maturity: maintenance
- function: llm-training, llm-inference, rag, data-generation, cli, chatbot
- domain: large-language-models, artificial-intelligence, machine-learning, developer-tools
- platform: python, cli
- tags: llm-alignment, synthetic-data, taxonomy, fine-tuning, model-tuning, community-contributions, linux, macos

## Member repositories
- instructlab/instructlab (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:40.396477+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 2026-08-30T04:37:53.397380+00:00, confidence not recorded.
  - readme: https://github.com/instructlab/instructlab (fetched 2026-08-28T04:04:40.396477+00:00, sha 363fbf062e75)
  - homepage: https://instructlab.ai (fetched 2026-08-29T11:50:21.069474+00:00, sha f76d8ad01d81)
  - site_page: https://docs.instructlab.ai/ (fetched 2026-08-29T11:50:21.072019+00:00, sha 4e73a027ab9f)
  - registry_pypi: https://pypi.org/pypi/instructlab/json (fetched 2026-08-29T11:50:21.074386+00:00, sha a67257d80714)
- Data as of 2026-08-30T08:39:29.467469+00:00.