← Back to community benchmarks

Laguna-S-2.1-oQ6e

M5 Max (40c) · 128 GB · 6bit · 2026-07-25
Performance
64k
tokens
801.3
PP tok/s
41.3
TG tok/s
81784
TTFT (ms)
90.1
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.5.3
macOS macOS 26.5.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 836.5 56.3 87.0 GB view
4k 1,011 55.3 87.2 GB view
16k 969.9 51.7 87.7 GB view
64k 801.3 41.3 90.1 GB current
Batching Results
Batch Size TG tok/s Speedup
56.3 1.00×
74.8 1.33×
100.2 1.78×
134.6 2.39×