← Back to community benchmarks
Qwen3.5-397B-A17B-8bit-gs32
Performance
16k
tokens
473.2
PP tok/s
22.3
TG tok/s
34621
TTFT (ms)
418.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 26.2
Context
16,384