← Back to community benchmarks

Qwen3.5-35B-A3B

M3 Max (30c) · 36 GB · 4bit · 2026-03-09
Performance
1k
tokens
1,049
PP tok/s
83.7
TG tok/s
976
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.2.6
macOS macOS 15.7.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,049 83.7 20.1 GB current
4k 1,139 79.5 20.2 GB view
Batching Results
Batch Size TG tok/s Speedup
83.7 1.00×
129.8 1.55×
189.1 2.26×