← Back to community benchmarks

Qwen3.5-122B-A10B-5bit

M5 Max (40c) · 128 GB · 5bit · 2026-03-27
Performance
16k
tokens
1,120
PP tok/s
41.8
TG tok/s
14627
TTFT (ms)
83.1
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.22
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 907.9 51.4 79.8 GB view
4k 1,363 48.6 81.3 GB view
16k 1,120 41.8 83.1 GB current
32k 974.4 35.3 85.5 GB view
64k 753.4 27.1 90.3 GB view
Batching Results
Batch Size TG tok/s Speedup
51.4 1.00×
74.8 1.46×
91.1 1.77×