GreyBrain Pulse · the daily AI briefing
Today on the
AI frontier.
Hundreds of new AI models ship every day. We surface the ones that matter and translate them into plain language. Tap any model for the brief.
Pick of the day
Today's model
🔥 23dQwen/Qwen3.8-27B
Qwen3.8-27B
--- libraryname: transformers license: apache-2.0 pipelinetag: image-text-to-text --- > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, TokenSpeed, etc. > [!Tip] > For users seeking managed, scalable inference without infrastructure maintenance, the official Qwen API service is provided by Qwen Cloud. …
Qwen3.8-27B
6.2M
downloads
14k
likes
Fresh drops
Trending now
In medicine
Models Hugging Face categorises under medical, healthcare or clinical tags. A category, not an endorsement.
Sunflower-Qwen-9B-medical-16bit
Image Text To Text · apache-2.0
Tap for the brief →ClinicalBERT
Fill Mask
Medical-NER
Token Classification · mit
Whisper-Medicalv1CNB1
Model
Mini-InternVL2-4B-DA-Medical
Image Text To Text · mit
Prompt of the day
Not a one-liner you could type yourself — an ambitious task that shows what AI can build for you: a hard clinical work-up, a patient handout, a stunning teaching image, a mini app. Tailored to today's model, with everything baked in. Copy it and run it.
Clinician prompt · Qwen3.8-27B
Untangle a complex med list
A structured interaction, dosing and deprescribing analysis with the reasoning shown — the kind of multi-step review that takes 20 minutes by hand.
Act as a clinical pharmacologist. A fictional 82-year-old woman (eGFR 28) is taking apixaban 5mg BD, digoxin 125mcg OD, furosemide 40mg OD, amitriptyline 25mg ON, ibuprofen 400mg TDS, metformin 1g BD, citalopram 20mg OD. Active problems: AF, CKD stage 4, osteoarthritis, depression, recurrent falls. (1) Flag every significant drug-drug and drug-disease interaction with its severity. (2) Tabulate renally-cleared drugs needing a dose change and the adjusted dose. (3) Apply STOPP/START for deprescribing candidates and missing indicated therapy. (4) Give a prioritized, week-by-week deprescribing and monitoring plan. Show your reasoning and name the rule behind each recommendation.
Synthetic case with a described image. Run it as-is, or attach a real de-identified image following your local safety policy.
Decision support only. Verify every recommendation against your formulary and the patient's full record.
Try it in a SpaceFrom the people who read this every day
Reading about AI is not the same as using it.
GreyBrain School of AI teaches doctors one clinical task at a time. A free textbook, a five-minute task each morning, and certificate programmes with IIHMR Bangalore when you want the credential.