← Back to community benchmarks

Qwen3.5-9B-DeepSeek-V4-Flash-oQ8-mtp

M5 Pro (20c) · 64 GB · 8bit · 2026-05-18
Performance
4k
tokens
1,411
PP tok/s
38.1
TG tok/s
2903
TTFT (ms)
10.9
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,066 38.5 10.2 GB view
4k 1,411 38.1 10.9 GB current
8k 1,471 38.9 11.4 GB view
16k 1,418 36.5 12.3 GB view