← Back to community benchmarks

Qwen3.5-122b-A10b-4bit

M4 Max (40c) · 128 GB · 4bit · 2026-03-10
Performance
4k
tokens
527.2
PP tok/s
47.3
TG tok/s
7770
TTFT (ms)
66.7
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 501.9 48.9 66.3 GB view
4k 527.2 47.3 66.7 GB current
8k 515.6 43.3 67.0 GB view
16k 483.8 37.7 67.7 GB view
32k 426.8 29.3 69.1 GB view
Batching Results
Batch Size TG tok/s Speedup
48.9 1.00×
77.5 1.58×
98.8 2.02×