So much of biomedical research rests on tacit knowledge—unwritten and invisible to LLMs—causing agents to underperform. The Biomni Open Know-How Catalogue bridges that gap: a curated library of human expertise that Biomni agents use intelligently on the fly, with dramatic gains.

This is one of the most effective ways to make your @ProjectBiomni agent super good! Teach your Biomni agent by adding your know-how. agent can intelligently follow your expertise on all related tasks. See our example about designing CRISPR sgRNAs: https://t.co/aZjIZGeJJ5

Excited to have @ProjectBiomni be part of Laude Slingshots program!

Biomni v0.0.8 is live! 🎉 🌟NEW: Gradio UI - Interactive web interface for Biomni agents, now open source! 🌟 Other features: 🧪ChEMBL, UniChem, QuickGO & DailyMed database integrations 📜Transcripformer & Arc State embeddings 🔎GPT-5 model support 👉Try it here:

Super exciting to be partnering with Nebius on our recent AI agents and benchmark work, building next generation infra in AI4Science, and wonderful to share progress on CRISPR-GPT, drug screening agent, and more next week. Join me with Artem from Nebius and talented student

🙌Biomni-R0-32B, the first RL agent model trained end-to-end for biology research, is now open-weight on @huggingface! To benchmark progress, we’re also releasing Biomni-Eval1 — 443 data points across 10 biomedical research tasks! 🔗 Data: https://t.co/TM0e5WTfkd 🔗 Model:

🧪Announcing @ProjectBiomni × @pylabrobot — a step toward an AI biologist that can think, design, and execute wet-lab experiments end-to-end. See Biomni in action below 👇 Super fun collab with @RetroBio_ @rickwierenga @serena2z! Learn more: https://t.co/VgqXbh8on2

🚀 Biomni v0.0.7 is live! New features: 📜 PDF export for agent chats 🦠 SOTA cell type transfer algorithms 🔬LazySlide pathology support 🔍Claude web search 🌟Bioimaging pipeline tools 🧬Gene conversion & ESM embeddings 🪴Glycoengineering capabilities Plus improved commercial

this guys calenders gotta be crazy

👋Excited to announce @Sagebio ✖️ @ProjectBiomni Biomni agent can now discover using 3.74 Petabytes of Sage data to enable deeper insights into rare disease, cancer, Alzheimer's, and more 🧬Learn more: https://t.co/aaddl8gTet

👋Announcing @ConsensusNLP X @ProjectBiomni Literature research underpins every biomedical breakthrough. We are excited to integrate the best literature engine with access to 200M+ peer-reviewed papers to Biomni for generating deeper and more factual biomedical insights!

🚀 Thrilled to share a preview of Biomni-R0 — we trained the first RL agent end-to-end for biomedical research. ➡️ nearly 2× stronger than its open-source base ➡️ >10% better than frontier closed-source models ➡️ Scalable path to hill climb to expert-level performance 🔗

🚀 Biomni v0.0.6 is live! New features: 🔬 https://t.co/VhhzBMuDVm integration 🧬 Copy Number Variant analysis tools added 🤖 Autonomous AI function generation 📚 Complete Sphinx documentation ⚡ Performance upgrades including centralized config management, fixed environment

🦠Microbiome analysis usually involves complex statistical workflows to identify disease-specific taxa and diversity patterns from thousands of microbial features. Input your microbiome CSV → get novel insights into microbial biomarkers, gut-disease associations, and testable

⚛️ Molecular docking usually involves complex software setup and structure preparation that can take hours. Input your protein target 🎯 + small molecule (SMILES) 😆 → get binding poses, affinity predictions, and interaction analysis in minutes using tools like AutoDock Vina

Biomni open source version 0.0.5 is now released! 🎉 Since our last release, we’ve seen amazing contributions from the community: 📊 New Data Sources - OpenFDA and Monarch databases added, panhumanpy for cell type annotation, DepMap data for CRISPR screens and cancer cell lines

🔗@ProjectBiomni now supports MCP! Now you can have a real research assistant to share analysis results with Slack/email, upload to GitHub, etc. Biomni tools are also exposed as MCP servers! Thanks to @serena2z Shibahara Takuma Try it out: https://t.co/kYuqgrT91L

🧪Rare diseases diagnosis requires extensive phenotype analysis, genetic variant interpretation, and literature review across multiple databases. Input patient symptoms (HPO terms) + genetic variants → get ranked differential diagnoses, pathogenicity assessments, and targeted

We're excited to collaborate with @ProjectBiomni to bring @scverse_team tools closer to a code-free future. This is an early-stage capability, and your feedback will shape what comes next. Try it out and help us build a more accessible analysis experience.

Thrilled to share @scverse_team x @ProjectBiomni! Biomni agent now supports 10 scverse core packages including scanpy, squidpy, scirpy, pertpy, etc! You can now use natural language to unlock complex single cell, spatial, and perturbation data analysis to generate novel