Ant Ling open-sources Ling-3.0-flash-Sante, a 124B healthcare MoE model

AIModelScope releases Ling-3.0-flash-Sante-fp8, an FP8 healthcare model with 124B total parameters and 5.1B active per token. It matches the BF16 model's 73.9 average across five healthcare benchmarks and supports a 256K context window, with native reasoning, tool calling, and SGLang or vLLM deployment on compatible hardware. The post says the model is for research use only, not clinical diagnosis or treatment.

AIAnt Ling has open-sourced Ling-3.0-flash-Sante, a healthcare-enhanced MoE model with 124B total parameters and 5.1B active parameters per token. The company says it performs among the best of similarly sized models on professional healthcare tasks and complex medical reasoning. Its weights are now available, with strengthened capabilities in deep research, multi-agent collaboration, and tool use.

AILing-3.0-flash-Sante is now open-source, a 124B-parameter healthcare mixture-of-experts model that activates 5.1B parameters per token. It scores 54.29 on HealthBench Professional, up 8.56 points, and ranks #2 overall in the team's evaluation. It reaches 76.92 on AFUMED-ComplexReasoning, 90.61 on AFUMED-Drug, and 94.21 on MedQA-USMLE.

AIAnt Ling has open-sourced Ling-3.0-flash-Sante, a healthcare-enhanced MoE model with 124B total parameters and 5.1B active parameters per token. The company says it performs among the best of similarly sized models on professional healthcare tasks and complex medical reasoning. Its weights are now available, with strengthened capabilities in deep research, multi-agent collaboration, and tool use.