← Back to community benchmarks

Qwen3.5-35B-A3B2bit

M1 Max (32c) · 64 GB · 2bit · 2026-04-18
Performance
32k
tokens
821.6
PP tok/s
42.0
TG tok/s
39882
TTFT (ms)
14.3
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.6
macOS macOS 26.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 767.3 74.5 11.3 GB view
4k 982.9 74.3 12.0 GB view
8k 998.6 70.6 12.3 GB view
16k 947.1 61.9 12.9 GB view
32k 821.6 42.0 14.3 GB current
64k 586.9 7.0 17.0 GB view