← Back to community benchmarks

Qwen3.6-35B-A3B

M5 (10c) · 32 GB · 4bit · 2026-04-18
Performance
16k
tokens
1,023
PP tok/s
49.0
TG tok/s
16011
TTFT (ms)
19.9
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.5.dev1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 820.5 61.9 18.2 GB view
4k 1,118 59.5 18.9 GB view
8k 1,120 55.8 19.2 GB view
16k 1,023 49.0 19.9 GB current
32k 817.7 38.0 21.2 GB view
Batching Results
Batch Size TG tok/s Speedup
61.9 1.00×
91.2 1.47×
108.6 1.75×