← Back to community benchmarks

Gemma-4-26B-A4B-Heretic

M3 Ultra (80c) · 512 GB · 8bit · 2026-05-12

Performance

1k

tokens

1,185

PP tok/s

77.9

TG tok/s

864

TTFT (ms)

27.2

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,185	77.9	27.2 GB	current
4k	1,858	75.9	27.9 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	77.9	1.00×
2×	122.2	1.57×
4×	173.5	2.23×