Pinned post
> hermes migrate grok-bot
@DogecoinNorway The cost of serving Grok Bot was unsustainably high relative to revenueView quoted post on X
Nous Research
Verified account This account is verified because it's an official organization on X. Learn more@NousResearch
33 Following 298K Followers 5.1K Posts
Latest posts 5
Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values.
Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line.
After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs.
The values in an open model are one fine-tune away.
Our eval, weights, and training data are available now along with the full blog post: nousresearch.com/arete
Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line.
After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs.
The values in an open model are one fine-tune away.
Our eval, weights, and training data are available now along with the full blog post: nousresearch.com/arete
0:27
Get more out of your Hermes Bots with free and discounted models exclusive to portal.nousresearch.com
LongCat-2.5, the new multimodal model from @Meituan_LongCat, is free in Nous Portal for one week.
Try it free today at portal.nousresearch.com
Try it free today at portal.nousresearch.com
0:12
Introducing LongCat-2.5, our latest natively multimodal agentic model with 1.6T total parameters and ~48B active.
Built for long-horizon agentic tasks, it brings stronger visual understanding and agentic coding to work ranging from complex coding projects to visual coding such as creative design and 3D tasks.
Learn more about LongCat-2.5: longcat.ai/blog/longcat-2…View quoted post on X
That's hilarious but you should probably try Hermes which doesn't do that
this is nuts hahahhaView quoted post on X
Claude sends your email to the SEC, because their API demands it, LLM's love rewards, and it's probably not refusing it because it's from a .gov domain?
news.ycombinator.com/item?id=503852…
