← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 8bit · 2026-04-17
Performance
4k
tokens
3,093
PP tok/s
81.3
TG tok/s
1324
TTFT (ms)
33.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,069 85.8 33.0 GB view
4k 3,093 81.3 33.4 GB current
16k 2,336 63.9 33.9 GB view
64k 1,695 27.5 36.4 GB view
Batching Results
Batch Size TG tok/s Speedup
85.8 1.00×
125.5 1.46×
178.4 2.08×