
webmaster @internetxstudio formerly @NousResearch
us east 1
Joined March 2017
- Tweets1.3K
- Following2.3K
- Followers266
- Likes14.6K
rishav retweeted
legalizing weed sounded like a great idea when i was a 20yr old college kid smoking joints on the weekends
not so much now that i’m a grown adult who does not partake and does not appreciate getting BLASTED with weed stench every time i step out into public
rishav retweeted
I do not support pacing frontier models. If I am going to die at the hands of killer AI, I want it to be American, not Chinese. Buy American, Die American.
rishav retweeted
The Ctrl+C mess on Linux came from a bad Microsoft decision. Mac put Cmd+Z/X/C/V on a dedicated modifier in 1984. Windows 3.1 in 1992 copies them onto Ctrl - adding a key to keyboards was hard.
Linux desktops copied the Windows shortcut. Problem: Ctrl+C has meant "kill this process" in every Unix terminal since the 1970s.
Something awkward but seems hard to change for the Linux desktops:
- Terminal still a must
- Ctrl C is not about copy in terminals
- To run an application, double click usually doesn't work
- Some applications don't work with Wayland
- cannot creating a desktop launcher of an application easily
- various package managers that get people lost
BTW, what has omarchy solved?
rishav retweeted
so you are telling me there’s a cpu, gpu, memory, and storage shortage and everyone is also going to start self hosting frontier open source models?
rishav retweeted
design engineer principles
vercel.com/design/engineer
rishav retweeted
Nous Research presents Hermes 4, our latest line of hybrid reasoning models.
hermes4.nousresearch.com
Hermes 4 builds on our legacy of user-aligned models with expanded test-time compute capabilities.
Special attention was given to making the models creative and interesting to interact with, unencumbered by censorship, and neutrally aligned while maintaining state of the art level math, coding, and reasoning performance for open weight models.
rishav retweeted
Measuring Thinking Efficiency in Reasoning Models: The Missing Benchmark
nousresearch.com/measuring-t…
We measured token usage across reasoning models: open models output 1.5-4x more tokens than closed models on identical tasks, but with huge variance depending on task type (up to 10x on simple questions).
This hidden cost often negates per-token pricing advantages. Token efficiency should become a primary target alongside accuracy benchmarks, especially considering non-reasoning use cases.
Read the thorough review of reasoning efficiency across the open and closed model landscape in our latest blog post in collaboration with our researcher in residence, Tim.
See more of their work here: github.com/cpldcpu/
rishav retweeted
Today marks a really big achievement for Nous, but also potentially the AI Landscape.
We have begun a decentralized pretraining run of what is basically a dense Deepseek - 40B parameters, over 20T tokens, with MLA for long context efficiency.
All checkpoints, unannealed, annealed, the dataset, everything will be opensourced live as the training goes on.
Check out the blog the Psyche team leads @DillonRolnick, @theemozilla and Ari wrote in the quote tweet to learn *a lot* more about the infrastructure.
Announcing the launch of Psyche
nousresearch.com/nous-psyche…
Nous Research is democratizing the development of Artificial Intelligence. Today, we’re embarking on our greatest effort to date to make that mission a reality: The Psyche Network
Psyche is a decentralized training network that makes it possible to bring the world’s compute together to train powerful AI, giving individuals and small communities access to the resources required to create new, interesting, and unique large scale models.
We are launching our testnet today with the pre-training of a 40B parameter LLM, a model powerful enough to serve as a foundation for future pursuits in open science. This run represents the largest pre-training run conducted over the internet to date, surpassing previous iterations that trained smaller models on much fewer data tokens.
rishav retweeted
Come hack with us!
Announcing the Nous RL Environments Hackathon in SF!
Create with Atropos, Nous' RL environments framework, and claim your stake of a $50,000 prize pool.
Partners - @xai @nvidia @nebiusai @SHACK15sf @akashnet @LambdaAPI @tensorstax and @runpod
May 18th. Sign up below 👇👇
rishav retweeted
FIVE MODELS OMG WE DROPPED FIVE MODELS
Replying to @NousResearch
We also release a collection of artifacts created using environments in Atropos, including one new dataset and five new models for tool calling, corporate fundamentals prediction and new, experimental personalities with RLAIF.
huggingface.co/collections/N…
On tool calling, our environment with GRPO improved DeepHermes' tool calling capabilities on Berkeley's Function Calling Benchmark by 2.4x and 5x on simple and parallel tool calls, respectively.
On corporate fundamentals prediction, the model went from ~25% accuracy on directional changes to 50%.
And on alignment with RLAIF, you will have to qualitatively try them for yourself.
rishav retweeted
Today we’re releasing our Inference API that serves Nous Research models. We heard your feedback, and built a simple system to make our language models more accessible to developers and researchers everywhere.
The initial release features two models - Hermes 3 Llama 70B and DeepHermes 3 8B Preview
To ensure a smooth rollout, we implemented a waitlist system at Nous Portal here: portal.nousresearch.com
- Access will be granted on a first-come, first-served basis
- Once granted access, you can create API keys and purchase credits
- This is an OpenAI-compatible completions and chat completions API
- Right now all accounts start off with $5.00 of free credits
Nous inference offerings will expand overtime; we welcome your ideas to help shape the future.
rishav retweeted
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:































