← Back to community benchmarks

Qwen3.5-35B-A3B5.5bit

M4 Pro (20c) · 48 GB · 5bit · 2026-03-10
Performance
8k
tokens
813.2
PP tok/s
64.4
TG tok/s
10074
TTFT (ms)
23.6
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 674.6 70.7 23.3 GB view
4k 822.2 69.3 23.4 GB view
8k 813.2 64.4 23.6 GB current
16k 738.4 55.7 24.0 GB view
Batching Results
Batch Size TG tok/s Speedup
70.7 1.00×
104.0 1.47×
121.0 1.71×
141.4 2.00×