← Back to community benchmarks

Qwen3.5-122B-A10B3.7bit-VL

M5 Max (40c) · 128 GB · 3bit · 2026-03-18
Performance
4k
tokens
1,399
PP tok/s
56.9
TG tok/s
2928
TTFT (ms)
55.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.19
macOS macOS 26.4
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,190 60.1 53.8 GB view
4k 1,399 56.9 55.3 GB current
8k 1,293 54.7 55.9 GB view
Batching Results
Batch Size TG tok/s Speedup
60.1 1.00×
88.9 1.48×
105.3 1.75×