← Back to community benchmarks

gemma-4-26B-A4B-it-oQ6

M4 Max (40c) · 64 GB · 6bit · 2026-05-06
Performance
128k
tokens
616.1
PP tok/s
24.2
TG tok/s
212757
TTFT (ms)
27.2
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,065 85.4 20.3 GB view
4k 1,304 82.2 21.0 GB view
8k 1,365 78.0 21.3 GB view
16k 1,288 72.4 21.6 GB view
32k 1,112 63.1 22.5 GB view
64k 940.4 49.0 23.7 GB view
128k 616.1 24.2 27.2 GB current
Batching Results
Batch Size TG tok/s Speedup
85.4 1.00×
122.8 1.44×
152.1 1.78×
180.0 2.11×