← Back to community benchmarks

KyleHessling1-Qwopus-GLM-18B-Healed

M3 Max (40c) · 64 GB · 4bit · 2026-05-04

Performance

4k

tokens

374.0

PP tok/s

36.4

TG tok/s

10951

TTFT (ms)

11.0

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	354.8	37.3	9.9 GB	view
4k	374.0	36.4	11.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	37.3	1.00×
2×	56.3	1.51×
4×	131.6	3.53×