← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (20c) · 64 GB · 4bit · 2026-05-01
Performance
32k
tokens
1,810
PP tok/s
49.1
TG tok/s
18108
TTFT (ms)
16.0
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,518 75.1 14.3 GB view
4k 1,955 71.2 14.9 GB view
8k 2,051 65.4 15.1 GB view
16k 1,972 59.7 15.6 GB view
32k 1,810 49.1 16.0 GB current
64k 1,417 36.0 17.7 GB view
128k 971.4 23.8 21.2 GB view