lumina gguf
by calcuis
Open source · 4k downloads · 25 likes
1.8
(25 reviews)ImageAPI & LocalAbout
Lumina GGUF is an AI model specialized in generating images from text descriptions, optimized to run efficiently even on modest hardware configurations thanks to its quantized version. It leverages an advanced architecture combining a high-performance text encoder and a VAE decoder to produce high-quality visuals in just a few seconds. The model stands out for its fast execution speed and compatibility with tools like ComfyUI, enabling seamless integration into creative workflows. Ideal for artists, developers, or enthusiasts looking to generate images without heavy technical constraints, it offers an accessible alternative to traditional solutions that demand more resources.
Documentation
gguf quantized version of lumina
- run it straight with
gguf-connector
Code
ggc l2
GGUF file(s) available. Select which one to use:
- lumina2-q2_k.gguf
- lumina2-q4_0.gguf
- lumina2-q8_0.gguf
Enter your choice (1 to 3): _
- opt a
gguffile in your current directory to interact with; nothing else - you will get the image output in few seconds even with the beginner level gpu
run it with gguf-node via comfyui
- drag lumina2 (opt anyone you like) to >
./ComfyUI/models/diffusion_models - drag gemma2-2b [2.32GB] and tokenizer [4.24MB] to >
./ComfyUI/models/text_encoders - drag pig [168MB] to >
./ComfyUI/models/vae
reference
- base model from alpha-vllm
- finetune model from neta-art
- gemma-2-2b (act as text encoder) from google
- flux.1-dev vae (act as vae decoder) from black-forest-labs
- comfyui from comfyanonymous
- pig architecture from connector
- gguf-node (pypi|repo|pack)
- gguf-connector (pypi)
Capabilities & Tags
ggufgguf-nodegguf-connectortext-to-imageendpoints_compatible
Links & Resources