# bigcode-project/starcoder

Home of StarCoder: fine-tuning & inference!

Repository: https://github.com/bigcode-project/starcoder
Canonical: https://ross.abutalabs.com/products/starcoder
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-02-27T02:05:57+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 87
- inputs: {"age_days": 1227, "days_push": 919, "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 7506, forks 524 (observed 2026-08-28T04:10:00.395549+00:00)

## What it is
StarCoder is a 15B-parameter code language model trained on 80+ programming languages, and this repository hosts the fine-tuning and inference code for it. It includes training pipelines (including a chat fine-tune), evaluation, and quickstart code for code generation via Hugging Face transformers.

## Use cases
- generate code completions with an open code LLM
- fine-tune a code language model on my own data
- run starcoder inference locally on a GPU
- build a coding assistant chatbot
- evaluate code generation models
- merge PEFT adapter layers into a base model

## When to choose
- you want an open, permissively licensed code generation model with training and inference code
- you need to fine-tune or adapt a code LLM with PEFT/LoRA
- you want a self-hosted alternative to proprietary code assistants

## When to avoid
- you need a maintained, actively updated project - the repo's last release was early 2024
- you lack GPU hardware for 15B-parameter inference
- you just want to call the model via an API without training code

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, llm-training, machine-learning
- domain: large-language-models, deep-learning, artificial-intelligence, developer-tools
- platform: python
- tags: code-llm, huggingface, fine-tuning, peft, starcoder, code-completion, code-generation, gpu, linux

## Member repositories
- bigcode-project/starcoder (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:00.395549+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 2026-08-29T17:37:30.824059+00:00, confidence not recorded.
  - readme: https://github.com/bigcode-project/starcoder (fetched 2026-08-28T04:10:00.395549+00:00, sha 6467b74c5169)
- Data as of 2026-08-30T08:39:29.467469+00:00.