Find out more about our tool on our GitHub page: https://t.co/InDMvepCRD

MedGemma is a really interesting model - very small, multimodal, open, and does quite well in out-of-distribution medical tasks compared to much larger models. Would love to see more work thinking about how to improve & deploy this sort of LLM to support medical professionals

The @GoogleDeepMind Gemma family of open models continues to grow! Today we launched T5Gemma a new take for encoder-decoder models and a new multimodal version of MedGemma with a specialized SigLip variant for healthcare. 🚀 T5Gemma: - Adapts Gemma 2 into flexible

Check out our state-of-the-art open weights MedGemma multimodal model for making sense of longitudinal EHR data as well as medical text and medical imaging data in various modalities (radiology, dermatology, pathology, ophthalmology, etc.) See the blog post linked below! ⬇️

🏥Introducing MedGemma, part 2, including: 🔥A 27B multimodal MedGemma 👀MedSigLIP, a lightweight image/text encoder for medical image retrieval/classification 📜A technical report with details Blog: https://t.co/UGgTZOGrOo Paper: https://t.co/LGhr9ai1Hl

The MedGemma technical report is out 👇 https://t.co/6ZxcKqFgMK

Introducing new models for research & development of health applications: MedGemma 27B Multimodal, for complex multimodal & longitudinal EHR interpretation, and MedSigLIP, a lightweight image & text encoder for classification, search, & related tasks. → https://t.co/I318jVmsYD

The MedGemma collection consists of our most capable open models for health AI development. Today we're announcing two new additions to the collection: MedGemma 27B multimodal and MedSigLIP.

Google released MedGemma on I/O'25 👏 > 4B and 27B instruction fine-tuned vision LMs and a 4B pre-trained vision LM for medicine > available with transformers from the get-go 🤗 they also released a cool demo for scan reading ⤵️

Exciting news from #GoogleIO! We’re introducing MedGemma, our most capable open model for multimodal medical text and image comprehension built on Gemma 3.

This week we announced ✨𝐇𝐞𝐚𝐥𝐭𝐡 𝐀𝐈 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐞𝐫 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧𝐬 (𝐇𝐀𝐈-𝐃𝐄𝐅)✨ which is a collection of open weight models to accelerate the development of health AI for real clinical use-cases.

Excited to share our work on ‘Collaboration between clinicians and vision–language models in radiology report generation’, published today @NatureMedicine (1/N). https://t.co/KIxBvwGp38

Our friends at Google Research used Colab to interface with their new models! Check out CT Foundation, a new medical foundation tool that helps you train models on CT scans in Colab. Read and sign up here for access:

Built on our work from ✨Med-Gemini✨, we're thrilled to unveil ⚡️⚡️CT Foundation⚡️⚡️, a novel AI endpoint that simplifies CT scans analysis!

Announcing CT Foundation, a new medical imaging embedding tool that accepts a computed tomography (CT) volume as input and returns a small, information-rich numerical embedding that can be used to rapidly train models. Learn more and try it out yourself → https://t.co/AFXAj5edTE

@cs50 Gambol is very much in beta (for example, it only works for 2-player situations at the moemnt), but I wanted to get something out there to see if it would be useful and interesting to people. Feel free to reach out with comments and questions!

I officially started Gambol over 4 years ago at the beginning of the pandemic and evolved it more recently with the latest advancements in GenAI. But really I started it 17 years ago as my final project in @CS50 which *may* have been a little ambitious for the time...

If you don't have a specific hand in mind or you want to do some studying, just click the Random button to pick an interesting post-flop situation.

Once Gambol has given you the GTO strategy, it will explain it using poker theory. Then you can ask follow-up questions, for example asking it to explain the poker terms it just used.

On the backend, Gambol has been trained just like Pluribus, the first superhuman multiplayer Texas Hold'em AI, using counterfactual regret minimization (CFR) to find the game theory optimal (GTO) strategy for a given situation.

Just in time for the World Series of Poker, I taught Gemini to play Texas Hold'em! Meet Gambol ( https://t.co/mYnHdrsCce), an AI coach for 6-max Texas Hold'em. Describe the hand including your hole cards, the action, and the board and Gambol will tell you the optimal strategy...