← CanIRunLocal · All hardware · All models
Yes. Qwen3.6 35B-A3B (MoE) fits in fast memory on M3 Max (40c GPU) 48GB at Q4_K_M (25.9 GB needed, 36.0 GB usable), about 80 tok/s (estimate).
Hardware: M3 Max (40c GPU) 48GB · Model: Qwen3.6 35B-A3B (MoE)
M3 Max (40c GPU) 48GB has 48 GB of unified memory and 400 GB/s of memory bandwidth. The checker leaves macOS a share of unified memory, so Ollama can use 36.0 GB. Qwen3.6 35B-A3B (MoE) (36B, 3B active, Alibaba) at Q4_K_M is a 24 GB download and needs 25.9 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 24 GB |
| Memory needed | 25.9 GB |
| Usable fast memory | 36.0 GB |
| Fit | fits in fast memory |
| Estimated speed | ~80 tok/s (estimate) |
| Command | ollama run qwen3.6:35b-a3b-q4_K_M |
Estimates, 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