https://t.co/h5eei3KQLH is a private and secure Chat GPT alternative. It can unlock knowledge from your confidential docs. Make your employees more productive. Your confidential data doesn't have to leave your organisation.

Grok 3 is shockingly capable in the handful of tests I've done so far. Not sure there's much point in keeping that $20/month ChatGPT subscription any longer. Love how freaking fast it is too! Wonder if that'll hold once rolled out to all?

We had _Spaces_ first! Just for the record :)

It's the little things. Adding the file upload feature to the Chrome extension. Oh did we forget to mention... We released a Chrome extension, in the Chrome store now!

I shared the following note with the OpenAI team today.

Chunking techniques for RAG: Fixed-length: Simple but rigid. Sentence-based: Natural but variable size. Paragraph-based: Logical but inconsistent lengths. Recursive chunking: Flexible but uses different logic. Semantic: Context-aware but computationally expensive. Sliding

💯 this

We released a bunch of updates to the backend in prep for something new coming out soon! 🎉 - Web API improvements and fixes - UI improvements as part of Streamlit upgrade (though moving towards getting off Streamlit) - Demo on Streamlit community cloud fixed.

@ThePeterMick Private ChatGPT/Perplexity for businesses @docqai

The improved version of the @docqai Chrome extension is coming along nicely. UI is powered by @chakra_ui. Creating that custom component for Space selection was 10x easier, faster, and more enjoyable to dev with Chakra (+ Copilot).

This OpenAI structured outputs feature, released the other day, is a big deal. It might not sound cool/sexy but it's important. Anybody building real AI products knows that _reliably_ binding LLM output to the UI and other parts of the software system is hard. >>

Boom I'm organising another Space tomorrow! We'll cover the latest AI/LLM news and talk about what each of us is working on. Join if interested! https://t.co/VpP2Xrr7Cj

Finetuning models shouldn't be your opening move! https://t.co/uvGY5QrMdM

StreamPot is now working 🫖 Plz try it. Plz roast it. Plz use it for all your video processing needs.

Azure OpenAI GPT-4o 🎉

Good suggestion by pundits to run on fewer chips. Turns out that made @GroqInc faster 🏎️🏎️🏎️

#OpenAI GPT-4o on #Azure is now available! It was released last week during the MS Build event. Interesting I didn't see any chat/noise about it on Twitter or LinkedIn. Google doesn't bring up much obvious reporting about it either.

Those building RAG solutions: have you tried HyDE query rewrites? what's your experience of its performance? if not great, what's working better? HyDE (Hypothetical Document Embeddings) paper:

Rewrote the core RAG logic in @docqai today to use declarative query pipelines in Llama Index. We were using the RAG fusion retriever + vector and BM25 search. Everything was wrapped into an inflexible box. So the work involved recreating that from composable pieces. >>

In April, there were four major open LLM releases: Mixtral, Llama 3, Phi-3, and OpenELM. In his latest article, @rasbt summarizes and reviews the latest LLM releases and covers everything from performance to available training information. Read more:

I’m starting to suspect that the target audience for @chatprd is non PMs and I think that PMs should think hard about what that means for the future of the role.