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

Mac 192 GB for local AI

Which image and video models run on an Apple Silicon Mac with 192 GB of unified memory, which file to download for each, and how much of the memory they need.

Apple192 GB unified · ~175.4 GB for the GPU · 800–800 GB/s · Apple SiliconData 2026-09-25
Unified memory192 GB
For the GPU~175.4 GB
Older macOS~154.6 GB
Bandwidth800–800
ChipsM2 Ultra
FP8 filesNo
BackendPyTorch MPS
Runs well49 of 49

Specs: support.apple.com

01What runs on it

ModelVerdictBest fileSizeNeeded
image models
FLUX.1 [dev]Runs well16-bit23.8 GB26.1 GB
FLUX.1 [schnell]Runs well16-bit23.8 GB26.1 GB
FLUX.1 Kontext [dev]Runs well16-bit23.8 GB26.4 GB
FLUX.1 Krea [dev]Runs well16-bit23.8 GB26.1 GB
FLUX.1 Fill [dev]Runs well16-bit23.8 GB26.4 GB
FLUX.2 [dev]Runs well16-bit64.4 GB67.7 GB
FLUX.2 [klein] 9BRuns well16-bit18.2 GB20.5 GB
FLUX.2 [klein] 4BRuns well16-bit7.8 GB9.6 GB
Krea 2 (Turbo)Runs well16-bit26.3 GB28.9 GB
Qwen-ImageRuns well16-bit40.9 GB43.7 GB
Qwen-Image-Edit (2511)Runs well16-bit40.9 GB43.9 GB
Qwen-Image 2.1Runs well16-bit14.2 GB16.5 GB
Z-Image TurboRuns well16-bit12.3 GB14.3 GB
Z-Image (base)Runs well16-bit12.3 GB14.3 GB
Ideogram 4Runs wellQ8_010.1 GB22.6 GB
Boogu-Image (Turbo)Runs well16-bit20.6 GB23.2 GB
ERNIE-Image (Turbo)Runs well16-bit16.1 GB18.4 GB
HiDream-O1-ImageRuns well16-bit16.4 GB19.2 GB
Mage-Flow (Microsoft)Runs well16-bit8.2 GB10.0 GB
Lumina Image 2.0Runs well16-bit5.2 GB7.0 GB
HiDream-I1 (Full)Runs well16-bit34.2 GB37.0 GB
HiDream-I1 (Dev)Runs well16-bit34.2 GB37.0 GB
Stable Diffusion 3.5 LargeRuns well16-bit16.5 GB18.8 GB
Stable Diffusion 3.5 MediumRuns well16-bit5.1 GB6.9 GB
Chroma1-HDRuns well16-bit17.8 GB20.1 GB
SDXL 1.0Runs well16-bit6.9 GB7.1 GB
Illustrious XL / Pony (SDXL anime)Runs well16-bit6.9 GB7.1 GB
Stable Diffusion 1.5Runs well16-bit2.1 GB3.3 GB
HunyuanImage 2.1Runs well16-bit34.9 GB38.2 GB
video models
Wan 2.1 T2V 14BRuns well16-bit28.6 GB32.9 GB
Wan 2.1 T2V 1.3BRuns well16-bit2.8 GB5.6 GB
Wan 2.1 I2V 14B 480PRuns well16-bit32.8 GB37.1 GB
Wan 2.1 I2V 14B 720PRuns well16-bit32.8 GB39.6 GB
Wan 2.1 VACE 14BRuns well16-bit34.7 GB39.5 GB
Wan 2.2 T2V A14BRuns well16-bit28.6 GB32.9 GB
Wan 2.2 I2V A14BRuns well16-bit28.6 GB32.9 GB
Wan 2.2 TI2V 5BRuns well16-bit10.0 GB13.8 GB
Wan 2.2 Animate 14BRuns well16-bit34.5 GB39.8 GB
Wan Animate 2 (14B)Runs well16-bit32.8 GB38.1 GB
Wan 2.2 S2V 14BRuns well16-bit32.6 GB37.4 GB
SCAIL-2 (character animation)Runs well16-bit32.8 GB38.1 GB
HunyuanVideo (13B, original)Runs well16-bit25.6 GB29.9 GB
LTX-Video 13B (0.9.8)Runs well16-bit28.6 GB32.9 GB
HunyuanVideo 1.5Runs well16-bit16.7 GB21.0 GB
LTX-2 (19B)Runs well16-bit37.8 GB42.6 GB
LTX-2.3 (22B)Runs well16-bit42.0 GB46.8 GB
LTX-2.5 (22B)Runs well16-bit42.0 GB46.8 GB
MiniMax H3 (33B)Runs well16-bit66.3 GB72.1 GB
MiniMax H3 PrunedRuns well16-bit40.2 GB46.0 GB

Calculated from real file sizes plus working memory. How the numbers work.

02Good to know

A Mac with 192 GB shares that memory between the CPU and the GPU. On current macOS the GPU may use about 175.4 GB of it by default (older macOS versions: about 154.6 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 192 GB: M2 Ultra (memory bandwidth 800–800 GB/s — the higher, the faster). Everything about Macs and local AI →

03Measured and reported results

No measured results yet. Send yours.