Jev-compatible public API
Social post about a public Jev-shaped API served by an open Qwen3.6-35B-A3B model so anyone can try typed decisions.
This summary is based on how curated lists describe the project; its README was not available to review.
Overview
This entry is a post on X describing a public API that mimics Jev's interface (state plus typed questions in, probabilities out) but is backed by an open-weight Qwen3.6-35B-A3B model. The stated aim is open research: letting anyone try the typed-decision interface without TypeSafe access.
It is relevant to people comparing open decision models with Jev or experimenting with Jev-style request shapes. Details such as endpoint, pricing, calibration and how option probabilities are derived from the open model are not available in the sources collected here.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Can an open LLM serve a Jev-compatible typed-decision API?(inferred) | — |
How it works
- Send a Jev-shaped request (state plus typed questions) to the public endpoint
- An open Qwen3.6-35B-A3B model produces probabilities for each allowed answer
- Use the typed answers as you would Jev's
Strengths
- Lets people try the typed-decision interface without a TypeSafe account
- Backed by an open-weight model
Caveats
- Only a short social post is available; no documentation collected
- Calibration and accuracy relative to Jev are unknown
At a glance
- Type
- Discussion
- Status
- discussion
- Runtime
- Hosted API
Use cases and integrations
How curated lists describe it
Open research: a public Jev-shaped API backed by an open Qwen3.6-35B-A3B model so anyone can try the typed-decision interface.
yibie/awesome-jev
Frequently asked questions
What is this?
A post announcing a public API with Jev's request shape, served by an open Qwen3.6-35B-A3B model.
Is it the same as Jev?
No. It copies the interface, but the model behind it is an open LLM, so quality and calibration may differ.
Does it need a TypeSafe key?
The sources do not say; it is described as a public API independent of TypeSafe.
Related projects
- TypeAR★ 946
TypeLLM: Jev-inspired type-safe decoding for existing autoregressive LLMs on SGLang, with schema-guaranteed typed outputs.
ProjectbetaPython llama.cpp-based C++ server that turns any LLM into a System One decision model, reading answers from next-token logits.
ProjectbetaC++MLX library that turns an existing local model into a decision engine by reading candidate logits, with no fine-tuning, on Apple Silicon.
ProjectprototypePythonPython library that turns an open LLM into a Jev-style classifier via constrained single-token choice indexes in one forward pass.
ProjectprototypePythonOpen Jev-style decision model built on Gemma 3 270M with a scoring head, scoring all options in one forward pass on CPU.
ProjectresearchPythonOfficial code for the FastJEV paper: training-free reduction of redundant candidate computation in multimodal JEV-style models.
ProjectresearchPython