🩺 We’re excited to share our most recent collaboration, published at @COLM_conf 2025, the result of joint work between outstanding researchers at the @UW, @CarnegieMellon, @LTIatCMU, @allen_ai, and @CPHAI_Dartmouth, in partnership with @LavitaAI. 📄 Link to paper:

@NEJM We gave this @NEJM medical case challenge to a few frontier AI models to see how they would perform. All models predicted "Fracture of the shaft of the ulna and dislocation of the radial head," consistent with the current majority vote. See the details below 👇

@arena @OpenAI GPT-5.2 by @OpenAI is now available on Medical Sphere. Come test your medical cases using any frontier AI model, for free!! 👇

GPT-5.2 by @OpenAI is now available on Medical Sphere. Come test your medical cases using any frontier AI model, for free!! 🩺🏥 PS: It’s fair to assume we’ll always have the most up-to-date models on the platform, even if we miss announcing a new release! 😁😀

Some takeaways from the State of AI paper by @OpenRouterAI and @a16z, a 100T-token study on LLMs’ usage, focusing specifically on the health category compared to other domains: 🔹Health is the most fragmented major category, no sub-use case exceeds 25%. User activity is

We gave a @NEJM medical case challenge to a few frontier AI models to see how they would perform. The LLM council’s consensus is B) Empyema necessitans 👇 🔗 Details: https://t.co/WMpbgbUntI

@lukner @schowardjd We recorded two short videos to show how you can easily create medical cases on our platform and test all frontier AI models (for free) in one place! Two simple steps: 1⃣ Create your medical case, our team will quickly review it for quality. 2⃣ Once approved, you can run any AI

@lukner This app from @MedicalSphereAI makes it convenient to run multiple AIs on the same case. You've probably already tried it out. https://t.co/0BMQGfwyNH

We gave a @NEJM medical case challenge to a few frontier AI models to see how they would perform. The LLM council’s consensus is B) Empyema necessitans 👇 🔗 Details: https://t.co/WMpbgbUntI

We gave a @NEJM medical case challenge to a few frontier AI models to see how they would perform. Among the models, only @xai's Grok 4.1 predicted Rocky Mountain spotted fever (RMSF) as the most likely diagnosis, consistent with the current majority vote. See the details below 👇

@elonmusk And you can test Grok 4.1 for free on your medical cases on the Medical Sphere! 👇🚀

🔗Link to the case

We gave a @NEJM medical case challenge to a few frontier AI models to see how they would perform. Among the models, only @xai's Grok 4.1 predicted Rocky Mountain spotted fever (RMSF) as the most likely diagnosis, consistent with the current majority vote. See the details below 👇

@EricTopol If anyone’s interested in a side-by-side comparison of Gemini 3 Pro, GPT-5.1, or any other model (for free), we’ve made it super easy on the Medical Sphere! 👇

@claudeai Claude Opus 4.5 is now available on the @MedicalSphereAI! Come test it on your medical cases, for free! 🫡👇

@DrDatta_AIIMS Nice work! If anyone’s also interested in testing Gemini 3, Grok 4.1, GPT-5.1, or any of the other models (for free!), check out Medical Sphere! 👇

@RadMasterclass Descriptions by Grok 4.1, Gemini 3 Pro, and GPT-5.1 👇 🔗 Details: https://t.co/6VrSQwIq7b

@DrPharmDMDTh Predictions by Gemini 3 Pro, Grok 4.1, GPT-5.1, and Claude Sonnet 4.5 👇 🔗 Details: https://t.co/d9HmJ3drqX

@pedsimaging We hid the description to see if AI models could describe what the MRI image shows. How did they do? 🤔

@radRounds AI model committee — GPT-5.1, Sonnet 4.5, Gemini 3, and Grok 4.1 — loud and clear says the answer is B) Adenomyomatosis! 🫡 🔗 Details: https://t.co/GcmYMrHFLR

@MayoClinicINFD Responses from several AI models on this case 👇 🔗 Details: https://t.co/Qa3lUgucjq