@simpsoka Making steering better. Sometimes you just need to give a small correction or tweak and 4 hours later the agent is reminding you that it should always commit and push before it reads the readme file.
@trq212 Does adaptive thinking with max in https://t.co/6nZVYGcWri now work to increase thinking even more beyond the highest level? Or does it work to regulate to the task?
@baylisswagner Yet somehow he’s been fine with Paxton winning the state in election after election for AG since 2015. Media already trying to do the Democrats work for them
@AlexBores lol let’s put aside the fact this would be blatantly unconstitutional on several grounds. You want to screw over your fellow New Yorkers who might be paid out and let citizens in every other state profit?
Doesn’t seem like a smart way to get re-elected
@Acyn lol how many hours did @mkraju stand out there trying to get this clip. Quite stupid when the numbers show there were overwhelmingly @KenPaxtonTX voters at the polls. Anything trying to show otherwise is misinformation
Not worried about limits yet. Would be great to ramp up the model so it’s usable for power users who are used to frontier models. There is too much detail ignoring to keep sanity while testing out Grok Build at the moment.
Thank you so much for all the feedback on the Grok Build Beta.
Some of you reported hitting limits quickly. Our team found areas to improve caching, so we've reset Grok Build usage limits for all accounts.
Please keep sharing feedback - the team is here to help.
@rudrank Does this mean @openai will lose their compute advantage over Claude after this week? I use both pro20x plans and for the most part they tend to burn about same tokens on the same work I do. If Codex loses 2x that would be horrendous for a 20x plan.