← Back to community benchmarks

gemma-4-26B-A4B-it

M4 (10c) · 32 GB · 4bit · 2026-05-29
Performance
16k
tokens
326.4
PP tok/s
24.8
TG tok/s
50194
TTFT (ms)
15.4
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.12
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 331.0 27.8 14.2 GB view
4k 352.8 28.0 14.9 GB view
8k 353.9 28.8 15.0 GB view
16k 326.4 24.8 15.4 GB current
32k 278.6 22.1 15.8 GB view
64k 226.0 17.0 17.3 GB view
Batching Results
Batch Size TG tok/s Speedup
27.8 1.00×
44.3 1.59×
53.5 1.92×
64.4 2.32×