Ling 3.0 Flash Sante is inclusionAI’s health- and medicine-focused language model, built as a fine-tune of Ling 3.0 Flash.
The model is designed for medical knowledge reasoning, clinical safety, evidence-based retrieval, and long-horizon medical workflows.
| Developer | inclusionAI |
|---|---|
| Model ID | inclusionai/ling-3.0-flash-sante |
| Release Date | September 4, 2026 |
| Context Window | 262,144 tokens (~262K) |
| Max Output Tokens | 32,768 tokens |
| Modalities | Text → Text |
| Capabilities | reasoning, streaming, tool/function calling |
Medical specialization
Targets health and medicine applications, including medical knowledge reasoning and clinical safety.
Evidence-based workflows
Supports evidence-based retrieval and complex medical workflows.
Reasoning and tools
Reasoning is enabled by default, and tool or function calling is supported.
General-purpose retention
Retains general reasoning, coding, and agentic capabilities from the Ling 3.0 Flash base model.
Text interaction
The model accepts text input and produces text output.
Medical question answering
Used for multi-step clinical question answering and medical question answering.
Literature summaries
Used to create summaries of medical literature.
Clinical documentation drafts
Used for drafting clinical documentation.
Medical retrieval workflows
Supports evidence-based retrieval and tool-based medical workflows.
Sources: https://person.run/docs/models/inclusionai-ling-3-0-flash-sante · https://models.darvareh.com/inclusionai/ling-3.0-flash-sante?lang=en · https://llmtr.com/en/models/inclusionai/ling-3.0-flash-sante
Medical knowledge reasoning, including answering complex medical questions and supporting evidence-based retrieval workflows.
Clinical decision-support tools, with outputs reviewed by qualified professionals rather than used as medical advice.
Medical literature search and health-focused assistants.
Long-horizon medical workflows such as multi-step clinical question answering.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "inclusionai/ling-3.0-flash-sante", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "inclusionai/ling-3.0-flash-sante", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"inclusionai/ling-3.0-flash-sante","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Ling 3.0 Flash Sante This page | Reasoning with tool calling | |||
| Finance-domain reasoning over long documents | ||||
| High-volume, zero-cost chat and reasoning tasks | ||||
| Fast, low-cost reasoning and tool use at scale |
It is a health- and medicine-focused language model designed for medical knowledge reasoning, clinical safety, evidence-based retrieval, and long-horizon medical tasks.
Yes. It is used for medical question answering, including complex and multi-step clinical questions.
Yes. The model is used for evidence-based retrieval and literature summaries.
Yes. It is used for clinical documentation drafts.
Yes. Reasoning is enabled by default, and the model is designed for medical knowledge reasoning and general reasoning tasks.
Yes. It supports function calling for tool-based workflows, and tool calling works whether reasoning is enabled or disabled.
Yes. It retains general agentic capabilities in addition to its medical specialization.
Yes. The model retains general coding capabilities alongside its health and medicine capabilities.
It accepts text input only.
Yes. It uses a mixture-of-experts architecture with 124 billion total parameters and approximately 5.1 billion active parameters per token.
The provider catalogue identifies streaming as a capability for this model.
