Can the Mac 48 GB run Mage-Flow (Microsoft)?
Yes. 16-bit fits on the Mac 48 GB with 30.2 GB to spare.
16-bit (8.2 GB) needs about 10.0 GB in use, so 30.2 GB of the 40.2 GB stays free: room for a larger image, a LoRA or a ControlNet.
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | mage_flow_bf16.safetensors 16-bit · Comfy-Org/Mage-Flow | models/diffusion_models | 8.2 GB | Download → |
| Text encoder | qwen3vl_4b_bf16.safetensors Qwen3-VL 4B BF16 · Comfy-Org/Mage-Flow | models/text_encoders | 8.9 GB | Download → |
| VAE | mage_flow_vae_bf16.safetensors Mage-Flow VAE · Comfy-Org/Mage-Flow | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 17.5 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 23.5 GB of your 48 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 Mage-Flow file on 40.2 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit ← SAFETENSORS · Comfy-Org | 8.2 GB | 10.0 GB | fits · 30.2 GB spare | the original weights | Hugging Face → |
“Needed” adds 1 GB of working memory for Mage-Flow and a 0.8 GB reserve. Sizes 2026-10-01. FP8 and INT8 files are left out on a Mac: Apple GPUs cannot compute FP8, so ComfyUI either fails to load them or converts them back to 16-bit, which saves no memory. Use a 16-bit or GGUF file.
03Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run Mage-Flow on a Mac 48 GB, send your time per image and peak VRAM and it will appear here, credited.
04About the Mac 48 GB
A Mac with 48 GB shares that memory between the CPU and the GPU. On current macOS the GPU may use about 40.2 GB of it by default (older macOS versions: about 38.7 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 48 GB: M3 Max, M4 Pro, M4 Max, M5 Pro, M5 Max (memory bandwidth 273–614 GB/s - the higher, the faster). Everything about Macs and local AI →
48 GB unified · ~40.2 GB for the GPU · 273–614 GB/s · Apple Silicon. Everything that runs on it →
05Questions
How much VRAM does Mage-Flow need?
Around 6.0 GB with the smallest file (INT8) and 6.0 GB with an 8-bit file (INT8). The full 16-bit file needs about 10.0 GB.
Which Mage-Flow file should I download for the Mac 48 GB?
16-bit (8.2 GB) from Comfy-Org/Mage-Flow.
How much do I need to download for Mage-Flow on the Mac 48 GB?
17.5 GB: mage_flow_bf16.safetensors, qwen3vl_4b_bf16.safetensors, mage_flow_vae_bf16.safetensors.