← Back to community benchmarks

Qwen3.5-397B-A17Bvlm

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-30
Performance
64k
tokens
378.5
PP tok/s
27.6
TG tok/s
173151
TTFT (ms)
220.7
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