← Back to community benchmarks

gemma-4-31B-it-oQ8

M1 Max (24c) · 64 GB · 8bit · 2026-05-25
Performance
16k
tokens
85.4
PP tok/s
7.6
TG tok/s
191923
TTFT (ms)
36.1
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 64 GB
GPU Cores 24
Software
oMLX v0.3.10
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 89.5 8.7 31.8 GB view
4k 89.1 8.5 34.5 GB view
8k 88.9 8.2 35.0 GB view
16k 85.4 7.6 36.1 GB current
32k 76.2 6.6 39.2 GB view
64k 66.0 4.8 45.1 GB view