← Back to community benchmarks

Qwen3.5-122b-A10b-4bit

M4 Max (40c) · 128 GB · 4bit · 2026-03-12
Performance
32k
tokens
460.9
PP tok/s
32.2
TG tok/s
71093
TTFT (ms)
69.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.8
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 528.8 52.1 66.3 GB view
4k 564.0 51.4 66.7 GB view
8k 553.5 46.6 67.0 GB view
16k 522.2 41.8 67.7 GB view
32k 460.9 32.2 69.1 GB current
Batching Results
Batch Size TG tok/s Speedup
52.1 1.00×
82.2 1.58×
111.0 2.13×