---
title: "Current average generation speeds for local DeepSeek-V4-Flash-Q2, highest to lowest:"
date: 2026-06-15
canonical: https://solmaz.io/x/2066415473265389857/
x_url: https://x.com/onusoz/status/2066415473265389857
license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
---

Current average generation speeds for local DeepSeek-V4-Flash-Q2, highest to lowest:

Mac Studio M3 Ultra:   32 tok/s
MacBook Pro M5 Max:   30 tok/s
Apple ??? M4 Max:   25 tok/s
MacBook Pro M3 Max:   24 tok/s
Mac Studio M2 Ultra:   22 tok/s
NVIDIA DGX Spark / GB10:   13 tok/s

It seems macs' higher memory bandwidth is contributing here, though I'm not sure if GB10 performance could be improved (I do hope so, I have one!)

*Quotes a post by @antirez (https://x.com/antirez/status/2066233392916525379); its text is omitted here because it is not covered by this site's license.*