← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Ultra (80c) · 512 GB · 8bit · 2026-03-21
Performance
16k
tokens
869.7
PP tok/s
34.4
TG tok/s
18840
TTFT (ms)
126.5
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.19
macOS macOS 15.7.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
16k 869.7 34.4 126.5 GB current
32k 745.3 31.1 128.9 GB view