← Back to community benchmarks

Qwen3.5-122B-A10B-oQ4-mtp

M5 Max (40c) · 128 GB · 4bit · 2026-05-18
Performance
4k
tokens
1,447
PP tok/s
54.1
TG tok/s
2830
TTFT (ms)
70.0
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 979.5 56.5 68.6 GB view
4k 1,447 54.1 70.0 GB current
Batching Results
Batch Size TG tok/s Speedup
56.5 1.00×
78.2 1.38×
105.2 1.86×