← Back to community benchmarks

Qwen3.5-VL-122B-A10B-4bitCRACK-X

M5 Max (40c) · 128 GB · 4bit · 2026-03-17
Performance
1k
tokens
871.4
PP tok/s
61.8
TG tok/s
1175
TTFT (ms)
66.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.2.14
macOS macOS 26.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 871.4 61.8 66.3 GB current
4k 944.5 58.9 66.7 GB view
8k 978.7 56.2 67.0 GB view
Batching Results
Batch Size TG tok/s Speedup
61.8 1.00×
89.1 1.44×
106.5 1.72×