For anyone that argues that reading speed is sufficient for generative AI, we present Google at reading speed… 😉

Thanks Mistral AI this is super helpful!
github.com/mistralai/mist…

We just released mistral-finetune, the official repo and guide on how to fine-tune Mistral open-source models using LoRA:
github.com/mistralai/mist…

Also released Mistral-7B-Instruct-v0.3 with support for function calling with Apache 2.0 license:
models.mistralcdn.com/mistral-7b-v0-…

Announcing `mistral-finetune`, the official repo and guide on how to fine-tune Mistral open-source models:  github.com/mistralai/mist…

A little over a month since udio launched and Drizzy has released a new single sampling their music 🤯🤯🤯

H/t to Avocado Papi 🥑 for creating the amazing track!!! 👑

1/ B2B AI apps are more focused than ever on owning the workflow 💪. This is the “Wave 2” we wrote about last year, where AI synthesizes info to show us less content. I wrote a follow up piece to dive into examples and discuss how this trend will continue!
a16z.com/owning-the-wor…

Congrats to Edo Liberty and the @Pinecone team on the GA of Pinecone serverless!

Pinecone has transformed the #VectorDatabase landscape — here's to their latest innovation 💪

'Did you use Scarlett Johansson's voice to train your chatbot?'

Is this why Open AI released Mac only?

Huge thanks Thomas Schulz Jacob Shamberger Solaris + AgentOps 🖇️ Wordware Logicrite Anon browserbase 🅱️

+ judges Mickey Friedman Rajko Radovanović @ ICLR 2024 Lenny Bogdonoff Joseph Nelson Paul Klein IV kwindla Austin Peirson Luis Hector Chavez M James Hill-Khurana

More AI? Follow Alex Reibman 🖇️

Lmao

😅

This is a must watch talk from Fei-Fei Li on spatial intelligence which is the new frontier for AI (both gen AI and agentic / robotics).

ted.com/talks/fei_fei_…

it turns out that 50% of successfully working in a distributed company is understanding this meme

“But soon after, it was revealed that the singer from the “BBL Drizzy” beat didn’t exist… The vocals, melody and instrumental of the sample were generated by Udio, an AI music startup founded by former Google Deep Mind engineers. […] his tongue-in-cheek diss became the first

[New blog] 📢 As a former PMs, Yoko and I have lived the grueling dance between designers and developers to settle on the latest iteration of the UI. Today with generative AI, builders get to dramatically shortened the journey from ideas to a fullblown application. 🧵

📊Want to add your AI Coder / Language Model to the SWE-bench leaderboard (swebench.com)?

SWE-bench official submission process is now available!

github.com/swe-bench/expe…

Simply add your model's predictions, execution logs, and agent agent trajectories to the repo.

Introducing Zero-Shot Tokenizer Transfer (ZeTT) ⚡

ZeTT frees language models from their tokenizer, allowing you to use any model with any tokenizer, with little or no extra training.

Super excited to (finally!) share the first project of my PhD🧵

And so by combining GPT-3, GPT-4, and GPT-4o, you can actually get all the capabilities of GPT-11

in case you're wondering what the GPT-4o singing demo might sound like with udio

[female singer, indie folk]