Check my GPU →

Lumina Image 2.0 on 16 GB of VRAM

Yes. 16-bit fits on 16 GB with 9.0 GB to spare. A 2.6B model from Alpha-VLLM and the base of the popular Neta Lumina anime fine-tunes. Small and quick.

Image model · 2.61BAny 16 GB GPUData 2026-10-01
Runs well
Yes, with room to spare.

16-bit (5.2 GB) needs about 7.0 GB in use, so 9.0 GB of the 16 GB stays free: room for a larger image, a LoRA or a ControlNet.

Best file16-bit
File size5.2 GB
VRAM needed~7.0 GB
System RAM32 GB+
Memory map · 16 GB GPU7.0 GB / 16 GB
08 GB16 GB
Weights 16-bit · 5.2 GBWorking memory · 1.0 GBReserve · 0.8 GBFree · 9.0 GB

01What to download

PartFileFolderSize
Modellumina_2_model_bf16.safetensors
16-bit · Comfy-Org/Lumina_Image_2.0_Repackaged
models/diffusion_models5.2 GBDownload →
Text encodergemma_2_2b_fp16.safetensors
Gemma 2 2B FP16 · Comfy-Org/Lumina_Image_2.0_Repackaged
models/text_encoders5.2 GBDownload →
VAEae.safetensors
FLUX.1 VAE (ae) · Comfy-Org/Lumina_Image_2.0_Repackaged
models/vae0.3 GBDownload →
Total download · keep about the same free on disk10.8 GB

Sizes from Hugging Face, 2026-10-01.

02Every file, on 16 GB

FileSizeNeededOn this cardQualityDownload
16-bit ←
SAFETENSORS · Comfy-Org
5.2 GB7.0 GBfits · 9.0 GB sparethe original weightsHugging Face →
Q8_0
GGUF · calcuis
2.8 GB4.6 GBfits · 11.4 GB sparepractically identical to the originalHugging Face →
Q6_K
GGUF · calcuis
2.1 GB3.9 GBfits · 12.1 GB sparevery close to the originalHugging Face →
Q5_K_M
GGUF · calcuis
1.8 GB3.6 GBfits · 12.4 GB spareclose; small differences in fine detailHugging Face →
Q4_K_M
GGUF · calcuis
1.5 GB3.3 GBfits · 12.7 GB sparegood; some loss in fine detail and textHugging Face →
Q3_K_M
GGUF · calcuis
1.1 GB2.9 GBfits · 13.1 GB sparenoticeable loss of detailHugging Face →
Q2_K
GGUF · calcuis
1.1 GB2.9 GBfits · 13.1 GB spareheavy loss; a last resortHugging Face →

“Needed” adds 1 GB of working memory for Lumina 2.0 and a 0.8 GB reserve. Sizes 2026-10-01.