BTW - we have had questions about why the email templates are short in length - typically around 250 words. It's a corporate sales tactic to create meaningful and succinct messages that the recipient can read easily on the phone with no scrolling. The majority of emails are read on a cell phone and long messages will typically be abandoned if the reader is in a rush. You'll get more attention by hooking them in a series of short messages versus a mini-novel.
Today, we release an early checkpoint of LFM2-24B-A2B, our largest LFM2 model. This sparse Mixture of Experts (MoE) model has 24 billion total parameters with 2 billion active per token, showing that the LFM2 architecture scales effectively to larger sizes.
Will someone LITERALLY think of the children?!
City Learns Flock Accessed Cameras in Children's Gymnastics Room as a Sales Pitch Demo, Renews Contract Anyway
#Flock #Cameras #Children #SalesPitch #Surveillance #Privacy #AI #Tech

TL;DR: MCP tool definitions can burn 55,000+ tokens before an agent processes a single user message. We built the Apideck CLI as an AI-agent interface instead:an ~80-token agent prompt replaces tens of thousands of tokens of schema, with progressive disclosure via `--help` and structural safety baked into the binary. Any agent that can run shell commands can use it. No protocol support required.
However, I can't help but feel sceptical about the article posted. It sounds too much like a #salesPitch. I'm especially bothered by point 4 โ #cloudFirst. #CloudServices have plenty of sensible #useCases. However, for many businesses, the number drops radically once #BusinessObjectives are considered. #DataStorage is often best done #onprem, in many cases, purely from a financial perspective [1]. Google seems to hint at this approach introducing security issues. /2
[1] https://thenewstack.io/cloud-vs-on-prem-comparing-long-term-costs/