← Back to community benchmarks

Qwen3.5-9B

M1 Max (32c) · 32 GB · 8bit · 2026-03-09
Performance
64k
tokens
205.4
PP tok/s
14.8
TG tok/s
319131
TTFT (ms)
14.8
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 32 GB
GPU Cores 32
Software
oMLX v0.2.6
macOS macOS 26.2
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 260.1 33.8 10.7 GB view
4k 260.3 34.0 11.0 GB view
8k 272.4 32.0 11.2 GB view
16k 267.1 28.2 11.7 GB view
32k 233.2 20.8 12.7 GB view
64k 205.4 14.8 14.8 GB current
Batching Results
Batch Size TG tok/s Speedup
33.8 1.00×
42.8 1.27×
49.4 1.46×
47.2 1.40×