Lumina Image 2.0 on 6 GB of VRAM
Yes, but only just: Q8_0 leaves 1.4 GB free on 6 GB. A 2.6B model from Alpha-VLLM and the base of the popular Neta Lumina anime fine-tunes. Small and quick.
Runs well
Yes, just.
Q8_0 (2.8 GB) fits with only 1.4 GB to spare (about 4.6 GB of 6 GB).
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 How this is counted.
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | lumina2-q8_0.gguf Q8_0 · calcuis/lumina-gguf | models/unet | 2.8 GB | Download → |
| Text encoder | gemma_2_2b_fp16.safetensors Gemma 2 2B FP16 · Comfy-Org/Lumina_Image_2.0_Repackaged | models/text_encoders | 5.2 GB | Download → |
| VAE | ae.safetensors FLUX.1 VAE (ae) · Comfy-Org/Lumina_Image_2.0_Repackaged | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 8.3 GB | |||
Sizes from Hugging Face, 2026-10-01.
02Every file, on 6 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 5.2 GB | 7.0 GB | spills 1.0 GB | the original weights | Hugging Face → |
| Q8_0 ← GGUF · calcuis | 2.8 GB | 4.6 GB | fits · 1.4 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · calcuis | 2.1 GB | 3.9 GB | fits · 2.1 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · calcuis | 1.8 GB | 3.6 GB | fits · 2.4 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · calcuis | 1.5 GB | 3.3 GB | fits · 2.7 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · calcuis | 1.1 GB | 2.9 GB | fits · 3.1 GB spare | noticeable loss of detail | Hugging Face → |
| Q2_K GGUF · calcuis | 1.1 GB | 2.9 GB | fits · 3.1 GB spare | heavy loss; a last resort | Hugging Face → |
“Needed” adds 1 GB of working memory for Lumina 2.0 and a 0.8 GB reserve. Sizes 2026-10-01.
03The text encoder
At 5.2 GB even the smallest Gemma 2 2B file is too big for 6 GB, so it runs from system RAM. That slows down reading the prompt, not the picture. Sizes: 16-bit 5.2 GB. Why the encoder does not count against VRAM →