← Back to community benchmarks

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

M4 Max (40c) · 128 GB · 4bit · 2026-04-04
Performance
16k
tokens
1,386
PP tok/s
15.6
TG tok/s
11821
TTFT (ms)
71.0
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.2
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 517.1 57.4 67.5 GB view
4k 580.3 54.8 68.9 GB view
8k 573.0 51.1 69.5 GB view
16k 1,386 15.6 71.0 GB current