← Back to community benchmarks

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

M5 Max (40c) · 128 GB · 4bit · 2026-04-02
Performance
16k
tokens
1,434
PP tok/s
51.2
TG tok/s
11423
TTFT (ms)
68.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.0
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,050 63.1 65.6 GB view
4k 1,495 61.5 67.0 GB view
8k 1,531 58.5 67.7 GB view
16k 1,434 51.2 68.9 GB current