Ling-3.0-flash-Sante
InclusionAI's MIT-licensed 124B (5.1B active) hybrid-linear MoE tuned for medical reasoning and healthcare deep research.
Overview
Ling-3.0-flash-Sante is a healthcare-specialized model built on InclusionAI's Ling-3.0-flash, with 124B total parameters and 5.1B active per token. It uses a hybrid-linear mixture-of-experts design combining Kimi Delta Attention, gated MLA and sparse MoE layers, with training extended to 256K context. Thinking mode is on by default and tool calling is supported through compatible serving frameworks. The model card targets medical reasoning, healthcare research and tool-assisted workflows and states it is a research model, not a medical device. Weights are MIT-licensed; the model card gives benchmark scores only in figures.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- 124B total / 5.1B active parameters (hybrid-linear MoE)
- Kimi Delta Attention plus gated MLA layers
- 256K-token context
- Thinking mode on by default; tool calling supported
- MIT license
- Evaluated on medication-contraindication and response-safety tasks
Use cases
- Medical question answering and clinical-knowledge research
- Healthcare deep-research agents with tool use
- Base for domain fine-tuning in regulated settings (research use only)
Pricing
Free open weights under the MIT license on Hugging Face.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 19 Oct 2026
- Record updated
- 28 Oct 2026
- Record last checked
- 28 Oct 2026