been vibe coding for so long now that reading code snippets while reading articles and sdks make me nostalgic asf. can't deny that i miss coding by hand. the thrill of writing code, looking up documentation and articles, the debugging through hell. good times fr.
particularly excited to try and test the decisions api. the specs are insane, 128k context, multimodality and runs on luna. prices would obviously be astronomical compared to other rlcd models/apis. lets see if decisions api is worth it.
SpaceXAI's new unified π subscription is coming soon
One subscription gives you access to the entire ecosystem......Grok, Cursor, Grok Bot and π benefits all in one plan, with more benefits and access coming
@dark_coderz people are complaining about less limits but honestly not much of a caveat. if youre efficient, you could be coding all day and still have third party usage left at end of the month.
Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family.
Itβs a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
why isn't china ahead of the us in ai race? they are on a generational run of making smarter and cheaper models, they have probably the best power infra, the only caveat is compute, which shouldve been solveable given the engineering cap they have. am i missing smth here?