← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Pro (20c) · 64 GB · 8bit · 2026-05-04
Performance
32k
tokens
552.3
PP tok/s
36.8
TG tok/s
59331
TTFT (ms)
28.0
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 616.7 46.3 25.8 GB view
4k 700.5 45.3 26.4 GB view
8k 715.6 43.3 26.6 GB view
16k 627.4 40.9 27.1 GB view
32k 552.3 36.8 28.0 GB current
64k 472.1 21.8 29.2 GB view
128k 344.7 14.6 32.7 GB view
195k 252.3 11.5 36.5 GB view
Batching Results
Batch Size TG tok/s Speedup
46.3 1.00×
65.2 1.41×
75.3 1.63×
99.5 2.15×