flashprove
Further progress in provable inference will likely be an important part of improving trust in AI deployment. To that end, we're open-sourcing FlashProve’s code,
LiveUpdated 2026-11-06 13:30 UTC
AI research
AI research scores 51/100 on the AI Radar money board (#16), with 3103 posts and 79M views on X over the last 14 days.
It moved down 4 places in the last 24 hours. 83 new tools appeared in this market over the last 7 days, and 659 posts were published about it in the last 72 hours. AGK Intelligence also rates the room for a newcomer here 0.72/4.
Top AI research tools
moe-blogpost
A hands-on guide to training Mixture-of-Experts models
liberai.com
Technical blog
Science's Last Exam
Can AI do research, end to end? Science's Last Exam defines and measures whether an AI agent can carry a research goal around the whole loop of scientific…
science-last-exam
Science's Last Exam. Contribute to Gen-Verse/Science-Last-Exam development by creating an account on GitHub.
www-cdn.anthropic.com
Technical report
depth-hybrid-attention
Project overview & code
arXiv.org
We present EmbeddingGemma 2, an open multimodal embedding model which natively represents text, images, audio, video, and combinations thereof in a unified…
academy.dair.ai
Chat with Paper
lm_understand-logs
Complete logs of the experiments in the blog post 'LLMs can understand and write neural network weights' - RikiyaT/lm_understand-logs
alphaXiv
Now on @askalphaxiv
formal-math
Yes, it seems indeed the models even obtained the results on the same day. I haven't had a chance to look through the OAI claimed proof yet, but in the spirit o
visualrambling.space
An interactive visualization to understand how neural networks work
arXiv.org
Foundation models in changing environments can encounter unfamiliar domain knowledge that their original training did not cover. However, new domains rarely…
math
OpenAI paper
latent-sparse-attention
Code for Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention - Prefrontal-AI/latent-sparse-attention
arXiv.org
Agent harnesses repeatedly invoke reasoning after tool feedback, even when an earlier call has already fixed the next action. We introduce HarnessMM (Harness…
You don't need long context
Later tokens are costlier to generate than earlier tokens. On frontier open weight models, the disparity in cost may rise to as much as 7x. Through frequent…
harness2
Contribute to google-research/harness2 development by creating an account on GitHub.
arXiv.org
Agentic workflows are increasingly shifting towards long-horizon tasks, such as open-ended scientific work, where models benefit from long context windows.…
2611.06918
Join the discussion on this paper page
arXiv.org
Rufus-Air is an open and reproducible post-training recipe on GLM-4.5-Air-Base (106B-A12B), organized as a serial pipeline of eight stages: SFT, Reasoning RL,…
research.nvidia.com
Technical paper
metrictrail
Machine learning model. Download, explore files and run it on Hugging Face.
FAQ
Is there money in ai research?
AGK Intelligence rates buyer spend 1.82/4, search demand 1.16/4, sponsor budgets 1.58/4 and growth 2.99/4. Overall market score: 51/100 (#16 of 19).
What are the newest ai research tools?
flashprove, moe-blogpost, liberai.com, Science's Last Exam, science-last-exam. Updated 2026-11-06 13:30 UTC.
How is this ranked?
35% money (AGK Intelligence judgment), 25% traction on X, 20% momentum over 72 hours, 10% new tools, 10% room for newcomers. See methodology.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 13:30 UTC. Full method.