← Back to community benchmarks

Qwen3.5-122B-A10B

M4 Max (40c) · 128 GB · 4bit · 2026-03-26
Performance
8k
tokens
582.5
PP tok/s
49.9
TG tok/s
14063
TTFT (ms)
68.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.20
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 552.4 55.3 66.4 GB view
4k 595.8 52.3 67.9 GB view
8k 582.5 49.9 68.5 GB current
Batching Results
Batch Size TG tok/s Speedup
55.3 1.00×
85.8 1.55×
109.2 1.97×
134.6 2.43×