← Back to community benchmarks

Qwen3.5-9B

M3 Ultra (60c) · 256 GB · 4bit · 2026-03-21
Performance
8k
tokens
1,122
PP tok/s
95.8
TG tok/s
7303
TTFT (ms)
7.1
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,125 101.7 6.7 GB view
4k 1,137 102.9 7.0 GB view
8k 1,122 95.8 7.1 GB current
16k 1,083 78.2 7.6 GB view
Batching Results
Batch Size TG tok/s Speedup
101.7 1.00×
170.0 1.67×
204.6 2.01×