← Back to community benchmarks

Qwen3.5-9B-oQ2-mtp

M4 Max (40c) · 48 GB · 2bit · 2026-06-04
Performance
8k
tokens
854.5
PP tok/s
98.8
TG tok/s
9587
TTFT (ms)
5.7
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.4.1
macOS macOS 26.5.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 765.8 104.7 4.5 GB view
4k 850.3 103.6 5.1 GB view
8k 854.5 98.8 5.7 GB current
Batching Results
Batch Size TG tok/s Speedup
104.7 1.00×
140.7 1.34×
158.4 1.51×