# Stability-AI/StableLM

StableLM: Stability AI Language Models

Repository: https://github.com/Stability-AI/StableLM
Canonical: https://ross.abutalabs.com/products/stablelm
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Last push: 2024-04-08T19:26:02+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 88
- inputs: {"age_days": 1232, "days_push": 877, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 15684, forks 1003 (observed 2026-08-28T04:11:13.762055+00:00)

## What it is
StableLM is Stability AI's repository of open-weight decoder-only transformer language models, including the 3B-parameter StableLM-3B-4E1T trained on 4 trillion tokens. It hosts model checkpoints, training details, and technical reports rather than an inference framework.

## Use cases
- download open-weight small language model checkpoints
- run a 3B parameter LLM locally
- fine-tune an Apache-licensed base language model
- study multi-epoch pretraining on repeated tokens
- build a chatbot from an open language model

## When to choose
- you need a small, permissively licensed base LLM checkpoint to fine-tune or deploy
- you want to reproduce or study multi-epoch training regimes on a 3B model

## When to avoid
- you need a ready-made inference or serving framework - use vLLM or llama.cpp with these weights
- you need state-of-the-art frontier model quality
- you expect active development - releases stopped in 2024

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, llm-inference, machine-learning
- domain: large-language-models, deep-learning, artificial-intelligence
- platform: python, cross-platform
- tags: language-models, model-checkpoints, transformer, open-weights, stability-ai, gpu

## Member repositories
- Stability-AI/StableLM (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:13.762055+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 2026-08-29T17:05:34.962724+00:00, confidence not recorded.
  - readme: https://github.com/Stability-AI/StableLM (fetched 2026-08-28T04:11:13.762055+00:00, sha e0e38014f138)
- Data as of 2026-08-30T08:39:29.467469+00:00.