← CanIRunLocal · All hardware · All models
Yes. Hy-Memory 30B-A3B (MoE, memory) fits in fast memory on Mac Studio M5 Max (32c GPU) 36GB at Q4_K_M (19.6 GB needed, 27.0 GB usable), about 101 tok/s (estimate).
Hardware: Mac Studio M5 Max (32c GPU) 36GB · Model: Hy-Memory 30B-A3B (MoE, memory)
Mac Studio M5 Max (32c GPU) 36GB has 36 GB of unified memory and 460 GB/s of memory bandwidth. The checker leaves macOS a share of unified memory, so Ollama can use 27.0 GB. Hy-Memory 30B-A3B (MoE, memory) (30.06B, 3B active, Tencent) at Q4_K_M is a 18.24 GB download and needs 19.6 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 18.24 GB |
| Memory needed | 19.6 GB |
| Usable fast memory | 27.0 GB |
| Fit | fits in fast memory |
| Estimated speed | ~101 tok/s (estimate) |
| Command | ollama run hf.co/tencent/Hy-Memory-30B-A3B-GGUF: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