Deep Cogito (@deepcogito) 's Twitter Profile
Deep Cogito

@deepcogito

ID: 1907972046950424576

calendar_today04-04-2025 01:42:51

0 Tweet

2,2K Takipçi

2 Takip Edilen

Drishan Arora (@drishanarora) 's Twitter Profile Photo

Today, we are releasing 4 hybrid reasoning models of sizes 70B, 109B MoE, 405B, 671B MoE under open license. These are some of the strongest LLMs in the world, and serve as a proof of concept for a novel AI paradigm - iterative self-improvement (AI systems improving themselves).

Today, we are releasing 4 hybrid reasoning models of sizes 70B, 109B MoE, 405B, 671B MoE under open license.

These are some of the strongest LLMs in the world, and serve as a proof of concept for a novel AI paradigm - iterative self-improvement (AI systems improving themselves).
Dhruv Malrana (@drvdhruv) 's Twitter Profile Photo

We are putting out four frontier intelligence models today, each hitting heavy in its weight class. The large ones are among the smartest open models in the world right now. It has become increasingly clear that superintelligence is within reach. As we continue to iterate and

VentureBeat (@venturebeat) 's Twitter Profile Photo

Deep Cogito goes big, releasing 4 new open source hybrid reasoning models with self-improving 'intuition' venturebeat.com/ai/deep-cogito…

Drishan Arora (@drishanarora) 's Twitter Profile Photo

It is intuitively easy to understand why self play *can* work for LLMs, if we are able to provide a value function at intermediate steps (although not as clearly guaranteed as in two-player zero-sum games). In chess / go / poker, we have a reward associated with every next