← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 8bit · 2026-03-14
Performance
16k
tokens
1,842
PP tok/s
57.1
TG tok/s
8895
TTFT (ms)
37.0
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.10
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,823 66.6 36.2 GB view
4k 1,999 64.7 36.3 GB view
8k 1,976 61.8 36.5 GB view
16k 1,842 57.1 37.0 GB current
Batching Results
Batch Size TG tok/s Speedup
66.6 1.00×
103.8 1.56×
144.1 2.16×
189.7 2.85×