Dev Radar
Support
LiveUpdated 2026-10-31 14:22 UTC

Now 2 Sparks (256GB) against 1 M5U (256GB)

Now 2 Sparks (256GB) against 1 M5U (256GB) DeepSeek-V4.1-Flash-1017: a Mac Studio decodes code at 77.8 tok/s, 2 DGX…

This is a dev post classified by AGK Intelligence as Hosting & infra (a tool drop), kept by the Dev Radar because it carries real work, not commentary.

Now 2 Sparks (256GB) against 1 M5U (256GB) DeepSeek-V4.1-Flash-1017: a Mac Studio decodes code at 77.8 tok/s, 2 DGX Sparks get the first token 1.2x sooner at 32K 2 Sparks (256 GB), Mia's recipe: Decode, code: 72.1 tok/s Decode, prose: 65.2 tok/s 500 words: about 10.5 s First token, 32K: 15.4 s 1 at once: 57.6 tok/s 4 at once: 101.3 tok/s 8 at once: 99.5 tok/s Mac Studio M5 Ultra, Bob's recipe: Decode, code: 77.8 tok/s Decode, prose: 70.5 tok/s 500 words: about 9.6 s First token, 32K: 18.3 s 1 at once: 59.8 tok/s 4 at once: 84.2 tok/s 8 at once: 84.5 tok/s One model, each box on its maker's

Posted by Niklas (2.3k followers) 1 h ago · 9 likes · 840 views · view the original post on X. Kept by the Dev Radar as Hosting & infra.

More dev work like this

Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.2k posts from 5k X accounts over the last 21 days, 8.9k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-31 14:22 UTC. Full method.