1/ We had early access from @GoogleDeepMind to Gemini 3.5 Flash's native Computer Use. On Cua-Bench it posted the highest mean reward of any frontier model we tested - 0.267, on KiCad tasks no model fully solves. At Flash speed and cost.
Pretty cool to see a popular fork of Codex OSS repo with 60k stars out there with my team being top contributors (due to preserving history with the fork)
Keep building
github.com/openinterprete…
PSA: If Codex Desktop Computer Use is broken on macOS, try the open-source CUA driver from @trycua
My bundled Computer Use has been broken since May (“runtime app is missing” / transport closed). But standalone CUA works fine.
CUA:
github.com/trycua/cua
Codex issue:
we were genuinely impressed by Gemini 3.5 Flash on our benchmarks. it’s a fast, cost-effective option you can use with Cua Driver today. docs: cua.ai/docs/tutorials…
Hermes Agent + Computer Use by @trycua is pretty cool! Looking at Hermes interacting with apps and windows is mind blowing and a bit scary at the same time 😂
Here on Mac with MiniMax M3 and Reachy Mini.
I started seeing CUA everywhere and I really like using it in Hermes Agent, it feels like magic.
Recording a video soon for interaction with Reachy Mini!
1/ We had early access from @GoogleDeepMind to Gemini 3.5 Flash's native Computer Use. On Cua-Bench it posted the highest mean reward of any frontier model we tested - 0.267, on KiCad tasks no model fully solves. At Flash speed and cost.
heard Slack is popular these days. Please test this update we had, and be honest with your feedback! docs.factory.ai/integrations/s…
(Droid can also work alongside you on your machine, thanks to @trycua.)
Your software Factory should meet you where you are, and with the model you need.
When you work with Factory in @SlackHQ threads, you get to choose which model you use, and which computer Droid works in.
🐨⚡️ @trycua had early access from @GoogleDeepMind to Gemini 3.5 Flash's native Computer Use! On Cua-Bench it posted the highest mean reward of any frontier model they tested -- at Flash speed and cost.