← Back to community benchmarks

Qwen3.5-REAP-97B-A10B

M3 Max (30c) · 96 GB · 4bit · 2026-04-17
Performance
32k
tokens
199.6
PP tok/s
17.9
TG tok/s
164128
TTFT (ms)
57.8
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 96 GB
GPU Cores 30
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
4k 256.6 24.7 53.6 GB view
8k 244.1 23.1 54.2 GB view
16k 229.2 21.2 55.4 GB view
32k 199.6 17.9 57.8 GB current