← Back to community benchmarks

gemma-4-26B-A4B-it-oQ8

M4 Max (40c) · 128 GB · 8bit · 2026-06-05
Performance
16k
tokens
1,348
PP tok/s
68.2
TG tok/s
12152
TTFT (ms)
27.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 15.7.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,186 77.2 25.8 GB view
4k 1,392 74.7 26.3 GB view
8k 1,403 72.7 26.7 GB view
16k 1,348 68.2 27.1 GB current
32k 1,223 60.9 28.0 GB view
64k 1,018 50.1 30.0 GB view