← Back to community benchmarks

Laguna-S-2.1-oQ4e

M5 Max (40c) · 128 GB · 4bit · 2026-07-23
Performance
32k
tokens
786.7
PP tok/s
50.0
TG tok/s
41655
TTFT (ms)
65.8
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.5.3
macOS macOS 26.5.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 902.1 61.1 60.6 GB view
4k 1,051 55.3 60.9 GB view
8k 908.4 57.0 61.3 GB view
16k 869.7 54.4 62.8 GB view
32k 786.7 50.0 65.8 GB current
64k 734.5 41.4 71.8 GB view
128k 496.9 21.6 83.8 GB view
Batching Results
Batch Size TG tok/s Speedup
61.1 1.00×
61.6 1.01×
76.1 1.25×