My Dot checked on my Claude Code benchmark running on my Mac

@morganlinton

added Oct 3, 2026

Very cool, using my Dot to check on my benchmark running on Claude Code, on my Computer, totally seamless. I could quickly see my Dot becoming a core way I interface with everything on my computer. The more I use it, the more I feel like the differentiator is going to be both the model, and just how darn good OpenAI is at Computer Use, it really can just do it all right out of the box. So far very impressed, congrats to the whole team at OpenAI, and with unlimited usage in chat rn, I don't even have to bug @thsottiaux for resets 😅

First-hand result: Morgan gave his Dot a standing task to watch his Sol 6.1 vs Opus 5.5 benchmark; the Dot took a read-only look at the run in Claude on his Mac and reported back '14 of 23 Low-effort tasks are complete, with task 15 active... remaining runs and judging still need to finish before the Opus 5.5 comparison is ready.' Real screenshot of the Dot chat attached. Found via a curated DevDay/Dots monitor page that embeds first-hand X posts.