… which is a bit unintuitive because one would expect that part of the promise of client libraries would be insulating their users from breaking changes made to the JSON APIs that they wrap I have been burned enough times now: I don't trust vendor client libraries not to break
TOOLS
-
API Documentation: JSON Specs Over Client Libraries
By
–
Request for model API providers: please put JSON API docs (with example curl commands) somewhere prominent I always look for those first, before client libraries, because my experience is that client libraries are more likely to have backwards incompatible breaking changes
-
Sweagent vs Devin: Speed Trade-offs and Feature Gaps
By
–
i did end up trying sweagent. its definitely faster but also very very workflow specific to making pr’s for issues. needs a lot more work to have the iterative feedback flow of devin. opendevin seems to be closer, but doesnt have browsing yet https://t.co/NXXSggMdyi
— swyx 🐣 (@swyx) 14 avril 2024i did end up trying sweagent. its definitely faster but also very very workflow specific to making pr’s for issues. needs a lot more work to have the iterative feedback flow of devin. opendevin seems to be closer, but doesnt have browsing yet
-
End to End Latency vs Opus Weaviate Embeddings
By
–
whats end to end latency? vs just opus + weaviate + embeddings
-
Devin AI: Autonomous Task Assignment Over Raw Speed
By
–
On Devin: I have early access (I take no money from them or any other AI lab), and I find it really fascinating, if not there yet. I am not a coder, so I can’t speak to coding. But it is a new method for working with AI, one built around assigning work autonomously, not speed.
-
Using AI Tools Differently Than Chatbots Speed Matters
By
–
Disagree. You use it very differently than a chatbot. Faster would be better of course
-
Rock ‘n Roll Bard: AI Music Tools Make Creative Dream Possible
By
–
For about 5 years, I've had an audiobook concept I call Rock 'n Roll Bard. Kind of an audiobook with musical elements and a kickass soundtrack. AI music tools like Udio make me think that I might actually be able to produce this some day in the not-too-distant future.
-
Chris Messina’s Hashtag Invention and Twitter’s Initial Skepticism
By
–
True. It was such a brilliant tool, while it lasted. Invented by Chris Messina, a Twitter user, who was told by Twitter founders, it’s ‘too nerdy, won’t work’!
-
From Apple Notes to GitHub Issues: Capturing and Implementing Ideas
By
–
I use Apple notes for this about a dozen times a day, but I try to turn the more interesting notes into GitHub issues threads if I want to actually implement the ideas that I captured
-
llm.c Performance Optimization: Closing Gap with PyTorch
By
–
A few new CUDA hacker friends joined the effort and now llm.c is only 2X slower than PyTorch (fp32, forward pass) compared to 4 days ago, when it was at 4.2X slower The biggest improvements were:
– turn on TF32 (NVIDIA TensorFLoat-32) instead of FP32 for matmuls. This is a