# FlagAI-Open/FlagAI

FlagAI (Fast LArge-scale General AI models) is a fast, easy-to-use and extensible toolkit for large-scale model.

Repository: https://github.com/FlagAI-Open/FlagAI
Canonical: https://ross.abutalabs.com/products/flagai
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-13T03:53:45+00:00

## Health v2 (maintenance only)
Score: 64/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 92, release rhythm 8, longevity 100
- inputs: {"age_days": 1570, "days_push": 51, "days_rel": 664, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3869, forks 417 (observed 2026-08-28T04:08:27.734313+00:00)

## What it is
FlagAI is a Python toolkit for training, fine-tuning, and deploying large-scale AI models across NLP, CV, and vision-language tasks. It integrates parallel training backends like PyTorch, DeepSpeed, Megatron-LM, and BMTrain, and supports 30+ models including Aquila, AltCLIP, AltDiffusion, and GLM.

## Use cases
- fine-tune large language models on Chinese and English text
- train large models with data and model parallelism in under 10 lines of code
- download and use pretrained models like Aquila, AltCLIP, and AltDiffusion
- run few-shot learning with prompt-learning toolkits
- do text classification, information extraction, QA, summarization, and text generation
- generate images from text with bilingual diffusion models

## When to choose
- you need to train or fine-tune large models with distributed parallelism
- your focus is Chinese NLP tasks or bilingual text-to-image generation
- you want a single toolkit spanning NLP, CV, and vision-language models

## When to avoid
- you only need simple inference with mainstream Hugging Face models
- you require cutting-edge updates or broad community support, as development activity has slowed
- you work outside PyTorch-based GPU environments

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, machine-learning, deep-learning, nlp, image-processing, prompt-engineering
- domain: large-language-models, machine-learning, deep-learning, computer-vision
- platform: python
- tags: model-training, fine-tuning, multimodal, chinese-nlp, distributed-training, deepspeed, megatron-lm, bmtrain, text-to-image, few-shot-learning, natural-language-processing, gpu, linux

## Member repositories
- FlagAI-Open/FlagAI (main) score 64

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:27.734313+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 2026-08-29T18:25:49.242394+00:00, confidence not recorded.
  - readme: https://github.com/FlagAI-Open/FlagAI (fetched 2026-08-28T04:08:27.734313+00:00, sha e586ed4606c1)
  - registry_pypi: https://pypi.org/pypi/flagai/json (fetched 2026-08-29T09:20:05.275016+00:00, sha 593eef0c474b)
- Data as of 2026-08-30T08:39:29.467469+00:00.