Check my GPU →

ming_image_0.1_design_bf16.safetensors

On my RTX 5060 Ti 16 GB this file made a 1024x1024 image in 17.4 s, with the card peaking at 15.3 GB. It is the 16-bit file of Ming-Image 0.1 Design, 12.3 GB, and it goes in models/diffusion_models.

12.3 GBmodels/diffusion_modelsMing-Image
Fileming_image_0.1_design_bf16.safetensors
ModelMing-Image 0.1 Design (image model, 6.15B)
Precision16-bit
Size12.3 GB
VRAM to fitabout 14.6 GB
FolderComfyUI/models/diffusion_models/
Loader nodeLoad Diffusion Model (UNETLoader)
SourceComfy-Org/Ming-Image on Hugging Face

Download ming_image_0.1_design_bf16.safetensors (12.3 GB)

Where to put it

Put the file in ComfyUI/models/diffusion_models/ (portable version: ComfyUI_windows_portable/ComfyUI/models/diffusion_models/). Older guides say models/unet: that folder still works.

Which graphics cards it fits

VRAMThis fileBest file for that card (RTX 40/50)
6 GBNot practicalShort by 8.6 GBMing-Image-0.1-Design-Q3_K_M.gguf (5.1 GB)
8 GBOffload onlyShort by 6.6 GB: runs with part of it in system RAM, slowerMing-Image-0.1-Design-Q3_K_M.gguf (5.1 GB)
10 GBOffload onlyShort by 4.6 GB: runs with part of it in system RAM, slowerming_image_0.1_design_int8_convrot.safetensors (6.2 GB)
12 GBOffload onlyShort by 2.6 GB: runs with part of it in system RAM, slowerming_image_0.1_design_int8_convrot.safetensors (6.2 GB)
16 GBRuns wellFits in VRAM, 1.4 GB to sparethis file
20 GBRuns wellFits in VRAM, 5.4 GB to sparethis file
24 GBRuns wellFits in VRAM, 9.4 GB to sparethis file
32 GBRuns wellFits in VRAM, 17.4 GB to sparethis file
48 GBRuns wellFits in VRAM, 33.4 GB to sparethis file

Need = file × 1 + 1.5 GB for the picture and 0.8 GB for Windows and the browser - the same rule as the rest of the site.

Measured with this exact file

GPUTimeSizeStepsPeak VRAMSource
RTX 5060 Ti 16 GB17.4 s per image1024x10241215.3 GBmeasured by me, 2026-10-04

Time = sampling + VAE decode, after a warm-up run. Peak VRAM = memory in use on the whole card. How I measure.

What else you need

This file alone does not make a picture. With it you load:

RoleFileSizeFolder
Modelming_image_0.1_design_bf16.safetensors12.3 GBmodels/diffusion_models
Text encoderming_image_0.1_ling_mini_2.0_int8_convrot.safetensors19.5 GBmodels/text_encoders
VAEming_image_vae_bf16.safetensors0.3 GBmodels/vae

32.1 GB in total. Smaller text-encoder versions and other options are on the Ming-Image page.

Other files of Ming-Image

FilePrecisionSizeVRAM to fit
ming_image_0.1_design_bf16.safetensors16-bit12.3 GB14.6 GB
Ming-Image-0.1-Design-Q8_0.ggufQ8_07.3 GB9.6 GB
Ming-Image-0.1-Design-Q6_K.ggufQ6_K6.8 GB9.1 GB
ming_image_0.1_design_int8_convrot.safetensorsINT86.2 GB8.5 GB
Ming-Image-0.1-Design-Q5_K_M.ggufQ5_K_M6.2 GB8.5 GB
Ming-Image-0.1-Design-Q4_K_M.ggufQ4_K_M5.9 GB8.2 GB
Ming-Image-0.1-Design-Q3_K_M.ggufQ3_K_M5.1 GB7.4 GB
Ming-Image-0.1-Design-Q2_K.ggufQ2_K4.7 GB7.0 GB

Want the answer for your exact card? Check your GPU. Every file of every model: ComfyUI model files.

Size and source read from Hugging Face (2026-10-01).