AnotiaWang/awesome-decision-models

The AnotiaWang/awesome-decision-models project is a community-maintained, curated list of resources about decision models (also called System One models or typed decision models).

Instead of generating text like chat AIs, a decision model reads a state (text, JSON, or for some models images) plus a set of questions your code declares up front, and returns a probability for every possible answer——no generated prose. Questions come in three shapes:

• Noul – how likely a statement is true
• Choice – which one of your options is best
• Score – a level on an ordered scale (e.g., 1–5)

This format was introduced by TypeSafe AI with its Jev model and /v1/systemone API, and many models and tools listed here follow that same request shape.

The list is organized into sections:

• Hosted APIs – models you call over the internet (e.g., Jev, Microsoft-Decision-1)
• Open Models – downloadable weights you can run yourself (split into company/lab models and community models)
• Inference Techniques – ways to get decision behavior from existing models without training
• Runtimes & Platforms – serving infrastructure
• SDKs & Clients – code libraries to talk to these models
• Applications, Demos & Games – real-world uses and playable examples
• Agent Tools – integrations for AI agents
• Benchmarks & Evaluations – how models are tested
• Papers, Articles, and Related – research and background reading

The project is explicitly not affiliated with any model provider, and pull requests are welcome. You can browse the list on GitHub or the companion website to find models, APIs, tools, and research for building or using decision models.

If you want a quick reference: it's a directory of decision-model software and research, with linked explanations and comparisons——you read it to discover what exists, not to install it directly.