← Back to community benchmarks

Qwen3.5-35B-A3BIQ4_M

M5 Pro (20c) · 64 GB · 4bit · 2026-04-18
Performance
128k
tokens
837.5
PP tok/s
22.2
TG tok/s
156512
TTFT (ms)
33.3
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,454 55.8 22.2 GB view
4k 2,198 54.4 22.9 GB view
128k 837.5 22.2 33.3 GB current
195k 602.5 16.9 39.0 GB view
Batching Results
Batch Size TG tok/s Speedup
55.8 1.00×
68.8 1.23×
115.9 2.08×
184.0 3.30×