Datasheet for local AI49 models98 GPUsData read 2026-09-25
No adsNo tracking
Check my GPU →

Lumina Image 2.0 on 6 GB of VRAM

Image model · 2.61BAny 6 GB GPUData 2026-09-25
Runs well
Yes — and comfortably.

Download Q8_0 (2.8 GB). With the model's working memory it needs about 4.6 GB, leaving 1.4 GB spare on 6 GB. Quality: practically identical to the original.

Best fileQ8_0
File size2.8 GB
VRAM needed~4.6 GB
System RAM16 GB+
Memory map · 6 GB GPU4.6 GB / 6 GB
03 GB6 GB
Weights Q8_0 · 2.8 GBWorking memory · 1.0 GBReserve · 0.8 GBFree · 1.4 GB
calculated from real file sizes plus working memory. Not yet measured on this setup. How this works.

01What to download

PartFileFolderSize
Modellumina2-q8_0.gguf
Q8_0 · calcuis/lumina-gguf
models/unet2.8 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 disk8.3 GB

Sizes read from Hugging Face (2026-09-25). “Download” links start the file directly; the file name links are the same files the official ComfyUI workflows use.

System RAM: 16 GB or more. ComfyUI keeps the model file, the text encoder and the VAE in system RAM and moves them to the GPU as needed. With this set of files that is about 8.3 GB, plus roughly 6 GB for Windows, ComfyUI and a browser: 14.3 GB in total. With less RAM it still runs, but Windows starts swapping to disk and loading gets very slow. calculated

02Every file, on 6 GB

FileSizeNeededOn this cardQualityDownload
16-bit
SAFETENSORS · Comfy-Org
5.2 GB7.0 GBspills 1.0 GBthe original weightsHugging Face →
Q8_0 ←
GGUF · calcuis
2.8 GB4.6 GBfits · 1.4 GB sparepractically identical to the originalHugging Face →
Q6_K
GGUF · calcuis
2.1 GB3.9 GBfits · 2.1 GB sparevery close to the originalHugging Face →
Q5_K_M
GGUF · calcuis
1.8 GB3.6 GBfits · 2.4 GB spareclose; small differences in fine detailHugging Face →
Q4_K_M
GGUF · calcuis
1.5 GB3.3 GBfits · 2.7 GB sparegood; some loss in fine detail and textHugging Face →
Q3_K_M
GGUF · calcuis
1.1 GB2.9 GBfits · 3.1 GB sparenoticeable loss of detailHugging Face →
Q2_K
GGUF · calcuis
1.1 GB2.9 GBfits · 3.1 GB spareheavy loss; a last resortHugging Face →

Sizes from the Hugging Face file listing, read 2026-09-25. “Needed” = file + 1 GB working memory for this model + 0.8 GB kept free for the system.

03The text encoder

Gemma 2 2B: 5.2 GB as 16-bit. ComfyUI encodes the prompt first and can push the encoder out of VRAM before sampling, so it does not have to fit together with the model. On 6 GB the encoder itself is too big for VRAM, so let it run from system RAM: slower prompt encoding, same images.