← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 4bit · 2026-03-22
Performance
1k
tokens
1,996
PP tok/s
96.6
TG tok/s
513
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.20.dev3
macOS macOS 26.3.2
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,996 96.6 20.1 GB current
4k 2,344 90.4 21.0 GB view
8k 2,289 88.4 21.4 GB view
Batching Results
Batch Size TG tok/s Speedup
96.6 1.00×
153.6 1.59×
207.5 2.15×
262.7 2.72×