← Back to community benchmarks

gemma-4-26b-a4b-it

M5 (10c) · 32 GB · 4bit · 2026-05-06
Performance
16k
tokens
980.1
PP tok/s
33.0
TG tok/s
16716
TTFT (ms)
15.5
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 871.8 38.7 14.2 GB view
4k 1,002 37.2 14.9 GB view
8k 1,032 36.5 15.0 GB view
16k 980.1 33.0 15.5 GB current
32k 845.3 27.1 16.4 GB view
64k 655.5 17.5 17.7 GB view
Batching Results
Batch Size TG tok/s Speedup
38.7 1.00×
56.2 1.45×