jeez and GPT5 Pro on healthcare is capable of some advanced cardiology lol
@swyx
-
LiteLLM criticism and response abstraction layer question
By
–
wait this is why u guys dont like litellm? is there a “litellm of responses” yet?
-
Apple Intelligence Keynote Was Mostly Fabricated, Says Swyx
By
–
yeah so like this Apple Intelligence keynote was entirely faked lol. like 99% of it.
— swyx 🐣 (@swyx) 4 septembre 2025
i really fell for Apple's hype here, and that was a severe misstep. I really try to not be the ERMAGEHD THIS CHANGES EVERYTHING guy. Balanced takes with history's hindsight in mind here.
i'm… https://t.co/NAFsYIjtBNyeah so like this Apple Intelligence keynote was entirely faked lol. like 99% of it. i really fell for Apple's hype here, and that was a severe misstep. I really try to not be the ERMAGEHD THIS CHANGES EVERYTHING guy. Balanced takes with history's hindsight in mind here. i'm
-
Deep Agents Integration into LangChain 1.0 Question
By
–
is this basically folding deepagents into langchain 1.0?
-
Modular Models and Forgetting in Multi-Modal AI Systems
By
–
Q: "what about forgetting as you add modalities/languages" A: optimistic on modular models (me: like Apple i think but not sure what other research is relavant. see also
-
Tracking Sentiment Data Sources and Suggesting smol_ai
By
–
who do you track for sentiment data source? may i suggest the @smol_ai one?
-
Router needs 4-5x more parameter capacity
By
–
u get it there needs to be probably 4-5x more ooms of params in the router too
