← Back to community benchmarks

Qwen3.5-397B-A17B-4.551bit

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-30
Performance
64k
tokens
373.7
PP tok/s
25.5
TG tok/s
175370
TTFT (ms)
222.4
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
64k 373.7 25.5 222.4 GB current
128k 260.8 19.5 234.3 GB view