← CanIRunLocal · All hardware · All models
No. Qwen3-Coder 480B-A35B (MoE) does not fit on M2 Ultra 192GB: Q4_K_M needs 292.4 GB and 144.0 GB is usable.
Hardware: M2 Ultra 192GB · Model: Qwen3-Coder 480B-A35B (MoE)
M2 Ultra 192GB has 192 GB of unified memory and 800 GB/s of memory bandwidth. The checker leaves macOS a share of unified memory, so Ollama can use 144.0 GB. Qwen3-Coder 480B-A35B (MoE) (480B, 35B active, Alibaba) at Q4_K_M is a 290 GB download and needs 292.4 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 290 GB |
| Memory needed | 292.4 GB |
| Usable fast memory | 144.0 GB |
| Fit | does not fit |
| Estimated speed | no estimate, it does not fit |
| Command | ollama run qwen3-coder:480b |
Smallest hardware that runs it: 4x Mac Studio M5 Max (40c GPU) 128GB at Q4_K_M (fits in fast memory, 292.4 GB needed, 378.0 GB usable, about 26 tok/s (estimate)).
A smaller Alibaba model that fits this hardware: Qwen3-VL 235B-A22B (MoE) at Q4_K_M (fits if the Mac GPU limit is raised, 145.7 GB). Can a M2 Ultra 192GB run Qwen3-VL 235B-A22B (MoE)?
Open M2 Ultra 192GB 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