This thread gets me pretty excited about the future of GRID. Even the system we have set up right now (pre-defined workflows) works really well for a lot of queries (which we'll illustrate over the next few weeks). But there are some insane integrations coming (shout out to the

Didn’t even know the alphabet went this far

I'm getting hype, Sentient Chat is where it's at. Huge update coming soon😁

The boys Andy and robbie have been cooking with The Rollup. One of the first content channels I started watching to understand crypto, awesome to see them hit this milestone. The TBPN of crypto, 1M next🫡🫡

imma have to buy more credits, boutta rip an AI ad

I joined Sentient right as we wrapped up our fingerprinting research, and it was the first piece of technical information I had learn deeply. It was also one of the primary drivers I joined Sentient, because I was fascinated about the "push forward open-source AI in any way

Bearish (you can’t tell if that means buy or sell)

Things I wish I could experience for the first time again: - winning a game of Catan - watching Kung Fu Panda - jellying a layup - submitting an assignment at 11:59:59 PM - ice melts by drake - yamming a windmill poster in 2k14 mycareer - sinking a cup in die - the sense of

Elon's right to sue OpenAI and Apple imo. Open sourcing AI models isn't enough. The entire stack needs to be commoditized: data, compute, memory, workflows, pipelines, everything. If this doesn't happen, then only a few parties control the whole system. An AI monopoly will be

Open AGI Summits are undefeated, really enjoyed the one we put on last month in Berkeley focusing on AI. The Korea agenda is gonna be great with a big focus on robotics (the most interesting frontier of AI I think right now). Book your tickets now and ill see you guys there😁

The GRID is designed to turn frontier research into shared, open infrastructure. Open Deep Search (ODS) is one of our proudest additions: a modular retrieval + reasoning framework that shows how open-source systems can outperform proprietary stacks on real benchmarks. Fully

Things I wish I could experience for the first time pt.2: - watching Doc Hudson flip over a car - guessing the word in skribbl.io without any drawing - a dunkin donuts boston creme donut - en passanting against someone that doesn’t know about en passant - hitting

Led by Salah Al-Zubi, we are releasing Recursive Open Meta-Agent (ROMA), an open-source multi-agent framework for solving complex tasks. ROMA is inspired by the open-source AI writing project github.com/principia-ai/W… and builds upon state-of-the-art open-source projects:

ROMA is one of the best pieces of research we’ve put out. Industry best deep research according to the benchmarks, regardless of open OR closed-source. We’re doing everything to make open-source AI win, so we fully open-sourced it😁 Enjoy building🫡

This guy literally just told you how Nikita told you how the X algo works

Introducing ROMA, the open source deep research meta agent for Sentient Chat. It is the new leader on internet search benchmark like FRAMES and SealQA, outperforming previous SOTA performance reported by Kimi and other SOTA reasoning agents with search such as perplexity and