88Z AI, 24 June — Baichuan Intelligence and Tsinghua University jointly released the medical-enhanced large model Baichuan-M4, which ranked first globally in three HealthBench categories with an industry-low hallucination rate of 3.3%.
At a glance
- Baichuan-M4 ranked first globally in three HealthBench categories with a 3.3% hallucination rate.
- It features four core capabilities including deep consultation, enabling a shift from answering questions to clinical practice.
The update
Baichuan Intelligence and Tsinghua University have officially launched the new generation medical-enhanced large model, Baichuan-M4. The model achieved first place globally in three core categories of the authoritative HealthBench evaluation platform, with a hallucination rate of only 3.3%, the lowest in the industry. Baichuan-M4 features four core capabilities: deep active consultation, full-course memory, evidence anchoring, and Agent scheduling. By autonomously orchestrating diagnostic and treatment processes through Baichuan-Harness, the model has bridged the gap from merely answering questions to possessing actual clinical capabilities, significantly enhancing intelligent service levels in medical scenarios.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 百川智能Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.