← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Ultra (60c) · 256 GB · 8bit · 2026-03-17
Performance
8k
tokens
693.5
PP tok/s
38.8
TG tok/s
11813
TTFT (ms)
123.9
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.2.14
macOS macOS 26.3
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 651.0 42.1 123.2 GB view
4k 703.8 40.5 123.5 GB view
8k 693.5 38.8 123.9 GB current
Batching Results
Batch Size TG tok/s Speedup
42.1 1.00×
67.3 1.60×
90.4 2.15×
125.1 2.97×