community-verified · measured pool data
Can the M1 Max 64GB run gemma-4-26B-A4B-it?
Yes — community-measured at 49 tok/s (2 runs at 8-bit).
Verified single-stream runs on the M1 Max 64GB from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.
| Quantization | Median tok/s | TTFT (ms) | Peak VRAM | Max context tested | Engines |
|---|---|---|---|---|---|
| 8-bit | 49 | 462.2 | — | 2,048 | llama.cpp, ollama |