← Back to community benchmarks

qw3.6-27b

M5 Max (40c) · 64 GB · 4bit · 2026-04-30
Performance
16k
tokens
615.2
PP tok/s
23.5
TG tok/s
26632
TTFT (ms)
18.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.3.7
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 381.5 25.5 15.1 GB view
4k 539.4 33.2 16.5 GB view
8k 734.3 33.1 17.1 GB view
16k 615.2 23.5 18.4 GB current
32k 486.7 19.3 20.9 GB view
Batching Results
Batch Size TG tok/s Speedup
25.5 1.00×
37.0 1.45×
45.8 1.80×