Recent AI breakthroughs challenge the status quo narrative that only closed, mega labs have the ability to push the frontier of superintelligence.
Today we announce Nous Psyche built on @Solana - a cooperative training network for generative AI. Psyche coordinates heterogeneous hardware to join a run and train open-source models.
We retell the myth of Psyche — a mortal’s quest for retribution against divine odds:
You can now experiment with Psyche’s DisTrO-enabled training code on our GitHub, and the larger open-sourced distributed training stack will be released alongside testnet.
Nous Research announces the pre-training of a 15B parameter language model over the internet, using Nous DisTrO and heterogeneous hardware contributed by our partners at @Oracle, @LambdaAPI, @NorthernDataGrp, @CrusoeCloud, and the Andromeda Cluster.
This run presents a loss curve and convergence rate that meets or exceeds centralized training.
Our paper and code on DeMo, the foundational research that led to Nous DisTrO, is now available (linked below).
We harness both Nous DisTrO, our novel networking stack that reduces inter-GPU communication by up to 10,000x during pretraining, and the testnet code for Psyche, a decentralized network that builds on Nous DisTrO to autonomously coordinate compute for model training and more.
Psyche details coming soon.
DeMo was created in March 2024 by Bowen Peng (@bloc97_ ) and Jeffrey Quesnelle (@theemozilla) and has been published on arXiv in collaboration with Diederik P. Kingma (@dpkingma), co-founder of OpenAI and inventor of the Adam optimizer and VAEs.
Today we are launching the Forge Reasoning API Beta, an advancement in inference time scaling that can be applied to any model or a set of models, for a select group of people in our community.
The Forge Reasoning engine is capable of dramatically improving Hermes 70B to reach parity in some categories with OpenAI's o1 (full), at the cost of more inference compute.
The API is built upon three architectures developed at Nous:
1. Monte Carlo Tree Search (MCTS) 2. Chain of Code (CoC) 3. Mixture of Agents (MoA)
Together, these three techniques create a powerful reasoning system that outputs complex, flexible, and nuanced responses from LLMs. Elevating open-source ai to the level of frontier models has been a core principle of Nous since its inception.
We’re inviting a small group of beta users to try out the Forge Reasoning API over the next month. This inference technology requires battle testing and user feedback in order to determine what areas it uniquely excels at in the real world.