latent space explorer. musician. developer. architect.
Summit County, CO
Joined March 2019
- Tweets627
- Following360
- Followers1.7K
- Likes10.1K
A.J. retweeted
Gemini 2.0 Flash comes with native audio output, and it’s actually wild 🤯 we are working hard to roll this out quickly to more folks!
A.J. retweeted
LLMs play pictionary!
Playing with a new LLM benchmark: how good are they at drawing an SVG of a pelican on a bicycle?
Here's the difference between Claude 3.5 Sonnet new and Claude 3.5 Sonnet previous
ALT Claude 3.5 Sonnet (2024-06-20) - it's clearly a bicycle but the pelican has a pretty small beak
ALT Claude 3.5 Sonnet (2024-10-22) The pelican here has a better sized beak and is also colored an off-white
A.J. retweeted
One of our users, input_source, made a stunning animated 3d wave inspector. They did 95% of this, then I made a couple small adjustments -
"make this audio reactive"
"keep the controls where they are but add an option to hide them"
the buildup & crescendo ~00:50 floored me 🫨
A.J. retweeted
🦜🚎Parallel function calling for extraction
The newest OpenAI models allow you do parallel function calling
The main focus here as been on how this improves assistants, but this also makes extraction **significantly** easier for the following reasons:
🦟Less logic needed to create the LLM input (better developer experience)
🧮Less complicated function definitions to pass to the LLM (saves on tokens, less likely to confuse the LLM)
🗺️Less complicated output the LLM needs to produce (smaller chance the LLM messes up and outputs incorrect json)
✂️Less logic needed to parse the LLM output (better developer experience)
Blog: blog.langchain.dev/parralel/
Cookbook: github.com/langchain-ai/lang…
