← Back to community benchmarks

gemma-4-26B-A4B-it

M4 Pro (20c) · 48 GB · 4bit · 2026-06-01
Performance
16k
tokens
684.5
PP tok/s
52.8
TG tok/s
23935
TTFT (ms)
15.6
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
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 630.5 61.6 14.3 GB view
4k 708.9 58.9 14.9 GB view
8k 709.3 57.3 15.1 GB view
16k 684.5 52.8 15.6 GB current
32k 611.3 40.4 16.5 GB view
64k 429.4 33.9 18.5 GB view
Batching Results
Batch Size TG tok/s Speedup
61.6 1.00×
91.9 1.49×
107.1 1.74×