Can the Mac 32 GB run ERNIE-Image (Turbo)?
Yes. 16-bit fits on the Mac 32 GB with 8.4 GB to spare.
16-bit (16.1 GB) needs about 18.4 GB in use, so 8.4 GB of the 26.8 GB stays free: room for a larger image, a LoRA or a ControlNet.
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | ernie-image-turbo.safetensors 16-bit · Comfy-Org/ERNIE-Image | models/diffusion_models | 16.1 GB | Download → |
| Text encoder | ministral-3-3b.safetensors Ministral 3 3B · Comfy-Org/ERNIE-Image | models/text_encoders | 7.7 GB | Download → |
| VAE | flux2-vae.safetensors FLUX.2 VAE · Comfy-Org/ERNIE-Image | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 24.1 GB | |||
Only for some workflows: ERNIE prompt enhancer (optional) (6.9 GB). On a Mac, FP8 text encoders do not load, so a 16-bit one is listed. Sizes from Hugging Face, 2026-10-01.
Shared memory: about 30.1 GB of your 32 GB. On this machine the model, the text encoder, the VAE and the operating system all use the same memory. That leaves room, so nothing has to be swapped to disk. calculated
02Every ERNIE-Image file on 26.8 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit ← SAFETENSORS · Comfy-Org | 16.1 GB | 18.4 GB | fits · 8.4 GB spare | the original weights | Hugging Face → |
| Q8_0 GGUF · unsloth | 8.7 GB | 11.0 GB | fits · 15.8 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · unsloth | 6.8 GB | 9.1 GB | fits · 17.7 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · unsloth | 5.9 GB | 8.2 GB | fits · 18.6 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · unsloth | 5.0 GB | 7.3 GB | fits · 19.5 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · unsloth | 3.9 GB | 6.2 GB | fits · 20.6 GB spare | noticeable loss of detail | Hugging Face → |
| Q2_K GGUF · unsloth | 3.2 GB | 5.5 GB | fits · 21.3 GB spare | heavy loss; a last resort | Hugging Face → |
“Needed” adds 1.5 GB of working memory for ERNIE-Image and a 0.8 GB reserve. Sizes 2026-10-01.
03Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run ERNIE-Image on a Mac 32 GB, send your time per image and peak VRAM and it will appear here, credited.
04About the Mac 32 GB
A Mac with 32 GB shares that memory between the CPU and the GPU. On current macOS the GPU may use about 26.8 GB of it by default (older macOS versions: about 22.9 GB); that is the figure used here. ComfyUI can go past it, but macOS then starts compressing and swapping memory and everything slows down. ComfyUI runs on Apple GPUs through PyTorch's MPS backend. 16-bit and GGUF files work; FP8 and INT8 files do not save memory on a Mac, so they are skipped. Speed is the catch: even the fastest Macs are several times slower per image than a desktop RTX card. Chips sold with 32 GB: M1 Pro, M1 Max, M2 Pro, M2 Max, M4, M5, M6 (memory bandwidth 120–400 GB/s - the higher, the faster). Everything about Macs and local AI →
32 GB unified · ~26.8 GB for the GPU · 120–400 GB/s · Apple Silicon. Everything that runs on it →
05Questions
How much VRAM does ERNIE-Image need?
Around 5.5 GB with the smallest file (Q2_K) and 11.0 GB with an 8-bit file (Q8_0). The full 16-bit file needs about 18.4 GB.
Which ERNIE-Image file should I download for the Mac 32 GB?
16-bit (16.1 GB) from Comfy-Org/ERNIE-Image.
How much do I need to download for ERNIE-Image on the Mac 32 GB?
24.1 GB: ernie-image-turbo.safetensors, ministral-3-3b.safetensors, flux2-vae.safetensors.