# Alpha-VLLM/Lumina-mGPT-2.0

Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling

Repository: https://github.com/Alpha-VLLM/Lumina-mGPT-2.0
Canonical: https://ross.abutalabs.com/products/lumina-mgpt-20
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2025-11-03T07:50:38+00:00

## Health v2 (maintenance only)
Score: 42/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 50, release rhythm 35, longevity 36
- inputs: {"age_days": 517, "days_push": 303, "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 1084, forks 51 (observed 2026-08-28T04:03:31.461399+00:00)

## What it is
Lumina-mGPT 2.0 is a stand-alone decoder-only autoregressive model trained from scratch that unifies a broad range of image generation tasks, including text-to-image generation, image pair generation, subject-driven generation, multi-turn image editing, controllable generation, and dense prediction. The repository provides inference code, finetuning code, and pretrained 7B checkpoints on Hugging Face.

## Use cases
- generate images from text prompts
- edit images across multiple turns
- subject-driven image generation
- controllable image generation
- dense prediction from images
- generate image pairs
- finetune an autoregressive image model

## When to choose
- you want a single autoregressive model covering many image generation and editing tasks
- you need pretrained 7B checkpoints with inference and finetuning code
- you prefer a non-diffusion, decoder-only approach to image generation

## When to avoid
- you need a lightweight model for CPU-only or low-resource environments
- you only need a production-ready image generation API without research setup
- you require diffusion-based pipelines or ecosystem tooling around Stable Diffusion

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, llm-inference
- domain: artificial-intelligence, deep-learning, image-processing, large-language-models
- platform: python
- tags: text-to-image, autoregressive-model, image-generation, image-editing, diffusion-free, model-checkpoints, finetuning, gpu, linux

## Member repositories
- Alpha-VLLM/Lumina-mGPT-2.0 (main) score 42

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:31.461399+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 2026-08-30T06:49:54.683706+00:00, confidence not recorded.
  - readme: https://github.com/Alpha-VLLM/Lumina-mGPT-2.0 (fetched 2026-08-28T04:03:31.461399+00:00, sha 1965f1772222)
- Data as of 2026-08-30T08:39:29.467469+00:00.