← CanIRunLocal · All hardware · All models
No. Llama 4 Maverick 400B (MoE) does not fit on Mac Studio M5 Ultra (80c GPU) 96GB: Q4_K_M needs 247.7 GB and 72.0 GB is usable.
Hardware: Mac Studio M5 Ultra (80c GPU) 96GB · Model: Llama 4 Maverick 400B (MoE)
Mac Studio M5 Ultra (80c GPU) 96GB has 96 GB of unified memory and 1200 GB/s of memory bandwidth. Mac Studio M5 Ultra (64c GPU) 96GB has the same memory and bandwidth, so the fit numbers match; this page is Mac Studio M5 Ultra (80c GPU) 96GB. The checker leaves macOS a share of unified memory, so Ollama can use 72.0 GB. Llama 4 Maverick 400B (MoE) (400B, 17B active, Meta) at Q4_K_M is a 245 GB download and needs 247.7 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 245 GB |
| Memory needed | 247.7 GB |
| Usable fast memory | 72.0 GB |
| Fit | does not fit |
| Estimated speed | no estimate, it does not fit |
| Command | ollama run llama4:128x17b |
Smallest hardware that runs it: 4x Mac Studio M5 Max (40c GPU) 128GB at Q4_K_M (fits in fast memory, 247.7 GB needed, 378.0 GB usable, about 52 tok/s (estimate)).
A smaller Meta model that fits this hardware: Llama 4 Scout 109B (MoE) at Q4_K_M (fits in fast memory, 69.7 GB). Can a Mac Studio M5 Ultra (80c GPU) 96GB run Llama 4 Scout 109B (MoE)?
Open Mac Studio M5 Ultra (80c GPU) 96GB in the checkerEstimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-23.
A project by OnDevice Labs · llms.txt