← Back to community benchmarks

Qwen3.5-122B-A10B9bit

M3 Ultra (60c) · 256 GB · 8bit · 2026-03-10
Performance
1k
tokens
551.1
PP tok/s
43.1
TG tok/s
1858
TTFT (ms)
129.5
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.2.7
macOS macOS 26.3
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 551.1 43.1 129.5 GB current
4k 674.1 41.8 129.8 GB view
Batching Results
Batch Size TG tok/s Speedup
1× 43.1 1.00×
2× 65.4 1.52×
4× 85.4 1.98×