← Back to community benchmarks

gemma-4-26b-a4b-it-UD

M1 Max (24c) · 32 GB · 4bit · 2026-05-03
Performance
16k
tokens
368.3
PP tok/s
37.6
TG tok/s
44483
TTFT (ms)
16.0
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 32 GB
GPU Cores 24
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 350.0 45.6 14.7 GB view
4k 375.6 43.5 15.3 GB view
8k 377.5 41.4 15.5 GB view
16k 368.3 37.6 16.0 GB current
32k 345.6 31.2 16.9 GB view
64k 305.1 21.5 18.1 GB view
128k 241.9 13.8 21.7 GB view