d3
d3 is a 27B multimodal decision model that accepts text or JSON and can also process images and videos. It answers choice, yes/no, and scoring questions in one call, returning probabilities without generating text.
What it does
Use d3 models to classify inputs, answer yes-or-no questions, or rank options by probability.
- Availability
- Brought to Ollama in release v0.40.11
- Pricing model
- Free to try; Apache-2.0 license
- Deployment
- Available through Ollama
- Privacy
- Not established
- Licence
- Apache-2.0
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
9 claims- d3 answers choice, yes/no, and score questions in one call and returns probabilities for every answer. [1]
- **Many questions, one call:** Choice, Yes / No and Score questions about the same input are answered together, each from its own forward pass over the input, with a probability for every option.
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - Inputs can include text or JSON, images, and videos. [1]
| **Inputs** | Text or JSON, plus images and videos (several per request) |
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - The model is available at vllm-sr/d3 on Hugging Face and has a free hosted demo. [1]
- The model can be served with Transformers, vLLM, or vLLM Semantic Router.
- The model is licensed under Apache-2.0. [1]
**d3** is the 27B multimodal foundation decision model of Decision 3.0, the decision models of [vLLM Semantic Router](https://github.com/vllm-project/semantic-router).
Hugging Face. Quoted text is outside CC BY 4.0 and keeps its original rights. - d3 models accept text, structured state, or an image. [2]
Give a model text, structured state, or an image, and ask it to classify, answer yes or no, or score options with probabilities.
GitHub watchlist. Quoted text is outside CC BY 4.0 and keeps its original rights. - d3 models can classify, answer yes or no, or score options with probabilities. [2]
Give a model text, structured state, or an image, and ask it to classify, answer yes or no, or score options with probabilities.
GitHub watchlist. Quoted text is outside CC BY 4.0 and keeps its original rights. - Ollama brings vLLM Semantic Router's decision models to Ollama. [2]
d3 brings vLLM Semantic Router's decision models to Ollama.
GitHub watchlist. Quoted text is outside CC BY 4.0 and keeps its original rights. - The d3 family includes 2B, 4B, 9B, and 27B sizes. [2]
The family includes 2B, 4B, 9B, and 27B sizes.
GitHub watchlist. Quoted text is outside CC BY 4.0 and keeps its original rights.
Sources
2
1
Verified
2
GitHub watchlistOwner-controlled
https://github.com/ollama/ollama/releases/tag/v0.40.11
Retrieved 2026-10-28 00:02 UTC
Verified
Versions
8| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 8 | 2026-10-28 00:02 UTC | material change | capability, availability, deployment | Current |
| 7 | 2026-10-28 00:02 UTC | material change | capability, modality, access, deployment, pricing, availability | Current |
| 6 | 2026-10-26 01:12 UTC | material change | capability, modality, access, deployment, pricing, availability | Current |
| 5 | 2026-10-23 05:16 UTC | material change | capability, access, availability, pricing model, deployment | Current |
| 4 | 2026-10-15 12:29 UTC | material change | capability, access, deployment, pricing, availability, pricing model, deployment | Current |
| 3 | 2026-10-13 14:10 UTC | material change | access, availability, deployment | Current |
| 2 | 2026-10-12 15:08 UTC | material change | access, availability, deployment | Current |
| 1 | 2026-10-11 15:58 UTC | publication | None | Current |
In briefs
1- Generate editable slides and dashboards from a prompt Daily brief 2026-10-11
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.