← Back to community benchmarks

gemma-4-26B-A4B-it-oQ4

M5 (10c) · 32 GB · 4bit · 2026-04-29
Performance
16k
tokens
1,006
PP tok/s
22.5
TG tok/s
16282
TTFT (ms)
15.3
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.8rc1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 892.3 40.5 14.4 GB view
4k 1,008 38.3 14.9 GB view
8k 1,056 29.7 14.9 GB view
16k 1,006 22.5 15.3 GB current
32k 857.2 13.3 16.1 GB view
64k 662.7 9.5 17.8 GB view
Batching Results
Batch Size TG tok/s Speedup
40.5 1.00×
59.6 1.47×
71.5 1.77×