← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Max (40c) · 128 GB · 8bit · 2026-05-27
Performance
4k
tokens
1,354
PP tok/s
69.9
TG tok/s
3025
TTFT (ms)
26.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.11
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,108 72.5 25.8 GB view
4k 1,354 69.9 26.5 GB current
8k 1,356 68.4 26.6 GB view
16k 1,300 61.0 27.1 GB view
32k 1,166 54.4 28.0 GB view
64k 962.4 43.1 30.0 GB view
Batching Results
Batch Size TG tok/s Speedup
72.5 1.00×
104.6 1.44×
143.2 1.98×
184.7 2.55×