How to run Lumina-mGPT 2.0 locally

Alpha-VLLM (Shanghai AI Lab) · Autoregressive image gen + edit · 8B params · Apache-2.0 (commercial OK)

A stand-alone decoder-only autoregressive image model trained from scratch that unifies text-to-image, image-pair and subject-driven generation, multi-turn editing, controllable generation and dense prediction in one checkpoint.

What it costs to run — $0 markup

On your own machine — $0. Runs free if you have about 24 GB of VRAM (~16GB BF16 weights; realistic generation at high resolution wants a 24GB card, less with offload and smaller sizes.), via Transformers (Lumina-mGPT repo).
Rent a GPU — from $0.49/hr. Fits RTX A6000 48GB at the direct vendor price ($0 markup) — pay only for the minutes you generate.
Download the weights — free. Open weights at Alpha-VLLM/Lumina-mGPT-2.0.

Note: generative-media models are billed per image / per second / per minute on hosted services — not per token. Running locally or on your own rented GPU is usually far cheaper and keeps your data on your machine.

Key facts

Does	Text → image, Image editing, Subject driven generation, Controllable generation, Dense prediction
VRAM to run	~24 GB (~16GB BF16 weights; realistic generation at high resolution wants a 24GB card, less with offload and smaller sizes.)
Download	~16 GB
Parameters	8B
License	Apache-2.0 (commercial use OK)
Run with	Transformers (Lumina-mGPT repo)

Get Lumina-mGPT 2.0 on Hugging Face →

More multimodal / omni models

Qwen2.5-Omni-7B — Omni understanding + speech generation, ~24 GB VRAM
Janus-Pro-7B — Unified understanding + image gen, ~16 GB VRAM
OmniGen2 — Any to image (gen + edit + in context), ~17 GB VRAM
MiniCPM-o 2.6 — Omni understanding + speech generation, ~9.0 GB VRAM
BAGEL-7B-MoT — Unified understanding + image gen + edit, ~24 GB VRAM
Emu3.5 — Any to any world model (gen + edit), ~48 GB VRAM

Browse: all media models · chat / LLM models