Model Releases
Baichuan Releases Medical-Enhanced Reasoning Model Baichuan-M2
Original title:Today we introduced Baichuan-M2, a medically-enhanced reasoning model, which surpasses all open-sour...
AI summary
Today we launched Baichuan-M2, a medical-enhanced reasoning model that outperforms all open-source models, including gpt-oss-120b, on the #HealthBench benchmark. Try it here: https://huggingface.co/baichuan-inc/Baichuan-M2-32B #AIHealth
Key Takeaways
Today we launched Baichuan-M2,a medical-enhanced reasoning model that outperforms all open-source…
including gpt-oss-120b,on the #HealthBench benchmark.
Try it here: #AIHealth
Quick Facts
Published
August 11, 2025
Updated
November 6, 2026
AI score
71
Read the original
Read the full article at the original source
Why it matters
百川 M2 是首个明确主打医疗推理的 32B 模型,在 HealthBench 上压过 GPT-OSS,做医疗 AI 应用的可以直接上 HuggingFace 拉下来跑,垂类专精可能比通用巨兽更接地气。
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
Original article
· AIHOT entry
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Baichuan Releases Medical-Enhanced Reasoning Model Baichuan-M2. August 11, 2025. https://new.nat6.net/en/news/cmry2we7800gjroqa08di0u9e/
Orange Relay (pro.nat6.net) is a premium AI API relay — an excellent transfer station with competitive pricing, reliable quality and broad model coverage including Claude, GPT and Gemini.
Visit Orange Relay
OpenAI announced the launch of GPT-5 on its API platform, offering developers high reasoning performance and new…
OpenAI:官网动态(RSS · 排除企业/客户案例) Model Releases AI score 85
August 7, 2025
Cohere released Command A Reasoning, positioned for enterprise reasoning tasks. The company says it outperforms…
Cohere 产品与研究博客(网页) Model Releases AI score 61
August 21, 2025
OpenRouter has released GPT-5 on its platform; the model supports long context and is built for complex reasoning and…
OpenRouter:Announcements(RSS) Model Releases AI score 68
August 7, 2025
DeepSeek-V3.1 is officially released, with deepseek-chat and deepseek-reasoner corresponding to its non-thinking and…
DeepSeek:API 更新日志 Model Releases AI score 71
August 21, 2025
GPT-5 is now available on the OpenRouter platform, featuring long-context capabilities and designed for complex…
OpenRouter:Announcements(RSS) Model Releases AI score 68
August 7, 2025
DeepSeek-V3.1 was open-sourced as a hybrid reasoning model, allowing users to switch thinking mode with one click,…
公众号:DeepSeek(深度求索) Model Releases AI score 63
August 21, 2025
Previous article METR study update: algorithmic scoring overestimates AI agents' real software engineering capabilities
Next article this isn’t just a modeling problem. it’s also a benchmarking problem. spurious correlations are alw...