Skip to content
AI EmployeeSuper-AgentsAgent-to-AgentBenchmarksPricingBlog

Ant Group Opens LingSante-3.1, a 560B Medical Model

At a glanceQuick answers
What is LingSante-3.1-flash?
An open-weight medical model from Ant Group’s Ant Ling team: 560B total parameters, 25B active per token, a 1M-token context window, MIT license.
Is it better than Claude, GPT-6 and Gemini?
On some of Ant’s own benchmarks, yes; on others, no. No independent evaluation was published with the release.
Can it be used for diagnosis?
No. Its model card says it must not independently diagnose, prescribe, set treatment or make triage decisions.
Data illustration on a near-white ground: a large navy figure reads 560B with the words open medical model; an open padlock sits beside a clipboard with a stethoscope, with small cubes coming out of the lock.
Fig 0An open-weight medical model from Ant Group. Made by CellCog's image agent, running GPT Image 2.6.

Ant Group has released an open medical model that it says beats the top closed models on several health benchmarks. Its Ant Ling team announced LingSante-3.1-flash on X at 15:22 UTC on November 22, 2026: “an open-weight MoE model with 560B total parameters and 25B activations per token.” The team’s framing: “LingSante-3.1-flash moves beyond medical QA toward real-world healthcare problem.”

The weights are on Hugging Face under an MIT license. Here is what it is, where Ant’s own charts show it winning and losing, and what its model card warns.

On this page · 6 sectionsOpen
  1. What Ant released
  2. Where it leads, and where it trails
  3. What the model card warns
  4. What is missing
  5. What we are watching
  6. Sources
Key points5 · 4 min full read
  1. An open padlock with a cube inside.
    Ant Group’s Ant Ling team released LingSante-3.1-flash on November 22, 2026: an open-weight mixture-of-experts model with 560B total and 25B active parameters.
  2. A clipboard with a heart-rate line.
    The weights are on Hugging Face under an MIT license, with a 1M-token context window, aimed at medical reasoning and long clinical workflows.
  3. Two bars with a balance scale.
    On Ant’s own chart it leads Claude Opus 5.5, GPT-6 Astra and Gemini 4 Argon on HealthBench Professional, HLE with tools and BrowseComp, and edges GPT-6 Astra on EHR-Complex.
  4. Two bars with a balance scale.
    The same chart shows it trailing on MedXpertQA, MedEthicAlign, MCP-Atlas and PhysicianBench, and every score is the lab’s own.
  5. A chest X-ray frame with a magnifier.
    A second model, LingSante-3.1-air-VL (124B total, 5.5B active), reads medical images; the model card says neither should diagnose or prescribe on its own.

§ 01What Ant released

Item Value
Architecture Mixture of experts
Total parameters 560B
Active per token 25B
Context window 1M tokens
License MIT
Repository created November 19, 2026
Table 1LingSante-3.1-flash at a glance (Hugging Face model card, November 22, 2026)

Two hours later Ant added LingSante-3.1-air-VL, a multimodal model with 124B total and 5.5B active parameters for medical images, report writing and medical reasoning. The announcement had 246,868 views when we read it at 20:02 UTC.

§ 02Where it leads, and where it trails

Ant compares the model with DeepSeek, GLM, Kimi, Claude Opus 5.5, GPT-6 Astra and Gemini 4 Argon. Against the three US frontier models, the scores read off its chart split both ways:

Benchmark LingSante-3.1-flash Best of Claude, GPT-6 Astra, Gemini Result
HealthBench Professional 65.72 64.18 (GPT-6 Astra) Leads
HLE, text only, with tools 72.22 63.59 (GPT-6 Astra) Leads
BrowseComp, with context 94.10 91.50 (GPT-6 Astra) Leads
EHR-Complex 74.40 74.20 (GPT-6 Astra) Leads by 0.2
DiagnosisArena-MCQ 83.9 84.7 (GPT-6 Astra) Trails
PhysicianBench 57 58 (Gemini 4 Argon) Trails
MedXpertQA-Text 57.27 71.35 (Gemini 4 Argon) Trails
MedEthicAlign 80.85 88.08 (Claude Opus 5.5) Trails
MCP-Atlas 83.98 88.30 (Gemini 4 Argon) Trails
Table 2LingSante-3.1-flash against the best US frontier model on each benchmark (Ant’s chart)
HealthBench Professional, Ant's own chart (higher is better)Bar chart: LingSante-3.1-flash 65.72 highlighted, GPT-6 Astra 64.18, Claude Opus 5.5 61.19, Gemini 4 Argon 55.91, Kimi-K3.1-Pro 55.87, DeepSeek-V4.1-Flash 52.04LingSante65.72GPT-6 Astra64.18Opus 5.561.19Gemini Argon55.91Kimi K3.155.87DeepSeek V4.152.04HealthBench Professional, Ant's own chart (higher is better)Bar chart: LingSante-3.1-flash 65.72 highlighted, GPT-6 Astra 64.18, Claude Opus 5.5 61.19, Gemini 4 Argon 55.91, Kimi-K3.1-Pro 55.87, DeepSeek-V4.1-Flash 52.04LingSante65.72GPT-6 Astra64.18Opus 5.561.19Gemini Argon55.91Kimi K3.155.87DeepSeek V4.152.04
Fig 1HealthBench Professional, Ant's own chart (higher is better)

Ant’s own X post names PhysicianBench (57) as a strong result; on the same chart Gemini 4 Argon scores 58. The headline claim holds for the three benchmarks Ant names, HealthBench Professional, HLE and BrowseComp, and the margin on EHR-Complex is 0.2 points.

Animation: an isometric platform with an open padlock labelled Open weights, MIT, small cubes beside it, an orange tag reading Lab's own scores, six columns where four teal-capped ones are labelled Leads on 4 and two plain ones Trails on others, and a clipboard with a stop hand labelled Not a doctor; then the line Open and strong, not a doctor.
Fig 2Open weights, ahead on four of Ant's own charts, behind on others, and not a doctor

§ 03What the model card warns

Ant’s card is unusually direct about limits. “The model does not replace qualified healthcare professionals. It must not independently diagnose, prescribe, determine treatment, or make clinical triage decisions.” It adds: “Benchmark and simulated-environment results do not establish clinical effectiveness.” And on the 1M-token window: “Long-context support does not guarantee complete retrieval of every relevant fact.”

The card lists the intended uses as medical question answering and clinical document analysis with qualified human review.

§ 04What is missing

  • Independent scores. Every number above is Ant’s. Several benchmarks on its chart (AFUMED, MedEthicAlign, InclusionMed) are in a section Ant labels internal.
  • Running costs. The card gives serving setups for large GPU clusters, not a price or a hosted API.
  • Clinical evidence. The card itself says benchmark results do not establish clinical effectiveness.

§ 05What we are watching

  • Outside evaluations. HealthBench runs by groups other than Ant.
  • Hosted access. Whether OpenRouter or other providers list it, and at what price.
  • Rivals’ replies. Open medical models from other labs, and any update from Google, OpenAI or Anthropic on health benchmarks.

§ 06Sources

Frequently asked5 questions

Q1When was it released?

Ant Ling announced it on X at 15:22 UTC on November 22, 2026. The Hugging Face repository was created on November 19 and last updated on November 22.

Q2What license is it under?

MIT, according to its Hugging Face model card.

Q3What is LingSante-3.1-air-VL?

A smaller multimodal model, 124B total and 5.5B active parameters, for medical images, report writing and medical reasoning, announced at 17:26 UTC the same day.

Q4Are the benchmark scores independent?

No. All scores in this post come from Ant’s own charts. Several benchmarks, such as AFUMED and InclusionMed, are labelled by Ant as internal.

Q5Does CellCog use LingSante?

No. CellCog does not route to LingSante, and CellCog is not offered for patient health information.

Published 22 November 2026 All Category basics →