← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Max (32c) · 36 GB · 4bit · 2026-03-09
Performance
1k
tokens
1,228
PP tok/s
104.5
TG tok/s
834
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip M4 Max (32c)
Memory 36 GB
GPU Cores 32
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,228 104.5 20.1 GB current
4k 1,393 97.3 20.2 GB view
8k 1,373 91.9 20.4 GB view
Batching Results
Batch Size TG tok/s Speedup
104.5 1.00×
161.5 1.55×
229.0 2.19×