@thsottiaux / @OpenAIDevs
You started of this feature so keep it
we need this!
At least you could cut in weekly allowance to have the work done.
Right now it’s very annoying to have my work incomplete.
cc @theo / @maria_rcks, this time no using upto $800 of inference, this feature is much needed and helps in better UX
Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.
today we announce something truly special
introducing: Jesse v1
this model architecture is all new, from scratch, built and trained using public, popular datasets
it consumes 0 tokens, and does not have weights
it reasons, it believes, and it's very, very fast
They said that Jev has 0% hallucinations,
However, I found a common case where it does hallucinate.
I asked Jev how many r's in Strawberry, it said 2 with 88% confidence,
It's a classic o1 problem statement which we had,
Probably it has to do with CoT and hence reasoning
@typesafeai