osolmaz/switch
A framework for building decision-model harnesses for any kind of task. Your program offers the model a short list of next steps, and the model picks one at a time until it can answer.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
switch is a harness framework for decision models like Jev. It gives a decision model an agent loop for finite choices. Your code turns the state of a task into typed questions, such as which next step to take from a short list, and the model picks the answers. https://github.com/user-attachments/assets/a5d834cb-f55b-4f2c-96cf-61ab1ba1e432 A simple explainer for switch, also on YouTube A decision model answers typed questions with probabilities and writes no text, so it cannot generate the tool calls that a coding agent harness such as Pi relies on. switch generates the tool calls for it in code, and the model only picks one from that finite set. TypeSafe calls these models System One models, and its Jev was the first; llama.cpp serves open ones through the same API. switch puts what many people already write by hand, a decision-model API called over several steps, behind one abstract interface. You implement a…
Read the full README ↗ · Preview checked 2026-10-31T13:39:59.888Z
Inside the original README — Document outline
- switch
- Uses
- Concepts
- How a run works
- Install
- Examples
- Write a domain
- Write a harness
- Run, watch and replay
- License
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: MIT. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
A framework for building decision-model harnesses for any kind of task. Your program offers the model a short list of next steps, and the model picks one at a time until it can answer.
Repository facts
- Owner
- osolmaz
- Primary language
- TypeScript
- Stars
- 402
- Forks
- 27
- Open issues + pull requests
- 3
- License
- MIT
- Archived
- No
- Default branch
- main
- Created
- 2026-10-25T14:12:42.000Z
- Last push
- 2026-10-30T05:12:52.000Z
Topics and intended use
Owner-supplied topics: classifiers, decision-models, evals, llm-harness
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-11-01T12:17:46.653Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.