← Back to community benchmarks

gemma-4-26B-A4B-it-heretic

M4 (10c) · 32 GB · 4bit · 2026-05-09
Performance
16k
tokens
349.0
PP tok/s
28.9
TG tok/s
46946
TTFT (ms)
15.5
Peak mem (GB)
Hardware
Chip M4 (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 359.1 33.3 14.2 GB view
4k 369.9 32.2 14.9 GB view
8k 365.3 30.4 15.0 GB view
16k 349.0 28.9 15.5 GB current
32k 302.6 23.6 15.9 GB view
64k 236.1 16.6 17.7 GB view
Batching Results
Batch Size TG tok/s Speedup
33.3 1.00×
48.1 1.44×
53.6 1.61×