← Back to community benchmarks

gemma-4-26B-A4B-it

M4 Max (40c) · 128 GB · 4bit · 2026-05-22
Performance
32k
tokens
872.2
PP tok/s
73.5
TG tok/s
37571
TTFT (ms)
16.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.5
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 981.2 96.8 14.3 GB view
4k 903.8 97.9 14.9 GB view
8k 931.8 91.9 15.1 GB view
16k 949.9 85.7 15.6 GB view
32k 872.2 73.5 16.5 GB current
64k 626.2 48.4 18.5 GB view
Batching Results
Batch Size TG tok/s Speedup
96.8 1.00×
129.9 1.34×