← Back to community benchmarks

Qwen3.5-122B-A10B

M2 Ultra (60c) · 128 GB · 4bit · 2026-03-12
Performance
16k
tokens
352.6
PP tok/s
29.9
TG tok/s
46469
TTFT (ms)
67.7
Peak mem (GB)
Hardware
Chip M2 Ultra (60c)
Memory 128 GB
GPU Cores 60
Software
oMLX v0.2.8
macOS macOS 15.7.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 349.2 45.3 66.3 GB view
4k 367.5 42.2 66.7 GB view
8k 363.6 37.0 67.0 GB view
16k 352.6 29.9 67.7 GB current
Batching Results
Batch Size TG tok/s Speedup
45.3 1.00×
66.7 1.47×
83.5 1.84×
101.8 2.25×