← Back to community benchmarks

supergemma4-26b-uncensored4bit-v2

M4 Max (32c) · 36 GB · 4bit · 2026-04-17

Performance

4k

tokens

1,043

PP tok/s

79.1

TG tok/s

3927

TTFT (ms)

14.4

Peak mem (GB)

Hardware

Chip M4 Max (32c)

Memory 36 GB

GPU Cores 32

Software

oMLX v0.3.5.dev1

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	904.4	84.8	14.0 GB	view
4k	1,043	79.1	14.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	84.8	1.00×
2×	129.7	1.53×
4×	161.6	1.91×