Model reference · open weights
Lumina-mGPT-2.0-Omni is an open-weight image model from Alpha-VLLM. Lumina-mGPT-2.0-Omni (BF16) weighs 15.8 GB; the smallest configuration that runs it is RTX 4090 24 GB.
Summary of the Alpha-VLLM/Lumina-mGPT-2.0-Omni model card, 2026-10-01
What it is
| Released by | Alpha-VLLM |
|---|---|
| Released | 2025-08-02 |
| Parameters | 7.9B |
| VRAM | 15.8 GB for the weights |
What it runs on
| Card | One image | Memory |
|---|---|---|
| RTX 3060 12 GB … RTX 4060 Ti 16 GB | does not fit | |
| RTX 3090 24 GB | tight | 23.4 GB |
| RTX 4090 24 GB | tight | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
Lumina-mGPT 2.0 is a stand-alone, decoder-only autoregressive model, trained from scratch, that unifies a broad spectrum of image generation tasks, including text-to-image generation, image pair generation, subject-driven generation, multi-turn image editing, controllable generation, and dense prediction.
We provide the implementation of Lumina-mGPT 2.0, as well as sampling code, visit our GitHub.
If you find the provided code or models useful for your research, consider citing them as:
@article{xin2025lumina,
title={Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling},
author={Xin, Yi and Yan, Juncheng and Qin, Qi and Li, Zhen and Liu, Dongyang and Li, Shicheng and Huang, Victor Shea-Jay and Zhou, Yupeng and Zhang, Renrui and Zhuo, Le and others},
journal={arXiv preprint arXiv:2507.17801},
year={2025}
}
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel, then Templates → NetaYume Lumina Text to Image, and choose this file in its text encoder loader.
# on your rented machine (the ssh line is on its page in the console)
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the model (4.1 GB)
get Alpha-VLLM/Lumina-mGPT-2.0-Omni model-00003-of-00004.safetensors text_encoders
# what the template also loads “NetaYume Lumina Text to Image” (9.9 GB)
get duongve/NetaYume-Lumina-Image-2.0 NetaYumev35_pretrained_all_in_one.safetensors checkpoints
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST and PORT are your machine's, from its page in the console
ssh -L 8188:localhost:8188 dev@HOST -p PORT