← Back to community benchmarks

Qwen3.5-VLM-122B-A10B-oQ8p

M3 Ultra (80c) · 256 GB · 8bit · 2026-03-31
Performance
1k
tokens
747.7
PP tok/s
41.9
TG tok/s
1370
TTFT (ms)
122.4
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 256 GB
GPU Cores 80
Software
oMLX v0.3.0
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 747.7 41.9 122.4 GB current
4k 916.5 41.2 123.9 GB view
8k 919.3 38.8 124.5 GB view
16k 872.7 35.3 125.7 GB view
32k 764.5 29.8 128.1 GB view
64k 580.7 23.6 132.9 GB view
128k 366.9 15.6 142.6 GB view