← Back to community benchmarks

gemma-4-26B-A4B-it-uncensored-oQ5

M1 Max (32c) · 32 GB · 5bit · 2026-05-22

Performance

32k

tokens

670.1

PP tok/s

32.2

TG tok/s

48900

TTFT (ms)

19.6

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 32 GB

GPU Cores 32

Software

oMLX v0.3.9

macOS macOS 26.5

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	644.3	45.0	17.4 GB	view
4k	732.1	43.0	18.1 GB	view
8k	766.6	41.6	18.2 GB	view
16k	737.8	38.0	18.7 GB	view
32k	670.1	32.2	19.6 GB	current
64k	530.3	21.1	21.6 GB	view