← Back to community benchmarks

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

M3 Max (40c) · 128 GB · 4bit · 2026-05-14
Performance
16k
tokens
1,135
PP tok/s
64.0
TG tok/s
14434
TTFT (ms)
15.7
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 880.0 77.9 14.7 GB view
4k 1,118 74.2 15.4 GB view
8k 1,208 71.6 15.5 GB view
16k 1,135 64.0 15.7 GB current
32k 1,046 55.4 16.9 GB view
64k 851.3 42.5 18.1 GB view
Batching Results
Batch Size TG tok/s Speedup
77.9 1.00×
117.6 1.51×
154.1 1.98×
187.3 2.40×