Li Ruilin retweeted
Meet 𝗔𝗽𝗼𝗱𝗲𝘅 𝟭.𝟬 🔭 — a heavy-duty agent team for deep research, which sets the SOTA! The team searches the web, reasons over evidence, and writes reports where every claim is backed by an explicit 𝘦𝘷𝘪𝘥𝘦𝘯𝘤𝘦 𝘤𝘩𝘢𝘪𝘯, independently audited before delivery.
🌐 apodex.ai
Give MiroThinker a try! We’d love to hear your thoughts. ✨
Demo: dr.miromind.ai/
We just flipped the scaling narrative: Agentic Density > Parameter Count.
#MiroThinker 1.5 operationalizes Interactive Scaling—agents that seek evidence, iterate, and revise in real time (with a time-sensitive sandbox to avoid hindsight leakage).
Result: 30B hitting frontier-class agentic search, ~$0.07/query (≈20× cheaper vs 1T-class baselines).
Fully open source, read more: research.miromind.ai/blog/in…
Try: dr.miromind.ai
GH:github.com/MiroMindAI/MiroTh…
HF: huggingface.co/miromind-ai
134
Li Ruilin retweeted
Predicting is definitely one of the most valuable capabilities that we hope AI can help us achieve. Our model, MiroThinker-72B-Preview (releasing soon), just ranked #6 on FutureX and #1 among open-sourced models 🎉
Excited to share our latest work on brain dynamics foundation model -- Brain-JEPA: arxiv.org/abs/2409.19407!
🔥Thrilled to announce that Brain-JEPA, a groundbreaking foundation model for fMRI analysis, has been accepted to #NeurIPS2024 as a Spotlight! #AI4Neuroscience
Paper: arxiv.org/abs/2409.19407
Code: github.com/Eric-LRL/Brain-JE…
@HelenJuanZhou @sophiatinh @MNNDL_Lab @NeurIPSConf
155








