0160GitHub
Simple Jev
An open-source library that turns any Hugging Face model into a Jev-style endpoint.
Eugene Cheah and Featherless.ai wanted a Jev that is open source and can see images. Simple Jev takes any Hugging Face model and serves it as an API that answers with choices, scores and typed answers. The code is on GitHub, and it runs in production at Featherless.ai.
featherless-ai/simple-jevREADME ↗
# Simple Jev Project
Use compatible open models from huggingface, for structured classification and scoring, without training a separate classifier head.
Explore the demos, playground, and documentation at [simple-jev.featherless.ai](https://simple-jev.featherless.ai/).
Send shared context and a set of questions. Simple Jev reads the model's next-token logits for each question and builds a JSON response containing choices, rubric scores, or truth/support judgments. The model does not generate a JSON completion: the server constructs the response from the scores.
The current implementation runs locally with Hugging Face Transformers and PyTorch. Shared request validation, versioned prompt instructions, and response scoring live in the plain Python `common/` folder so other inference implementations can use the same rules.
## Try it today
Try the [public demo API](https://simple-jev-demo-api.featherless.ai/v1/) with **no login, API key, or authentication required**. The demo has a **2k-token context limit** and is **rate limited to 2 requests per second (2 RPS)**.
First, list the available models with `GET /v1/models`:
```bash
curl https://simple-jev-demo-api.featherless.ai/v1/models
```
The demo already serves Gemma as `featherless-ai/gemma-4-26B-A4B-classifier`. Try it directly with `POST /v1/classifier`, or use another model ID returned by the list:
```bash
curl https://simple-jev-demo-api.featherless.ai/v1/classifier \
-H 'Content-Type: application/json' \
--data-binary @- <<'JSON'
{
"model": "featherless-ai/gemma-4-26B-A4B-classifier",
"state": "Mia owns a red bicycle.",
"questions": {
"color": {
"type": "choice",
"instructions": "What color is Mia's bicycle?",
"criteria": {"red": null, "blue": null}
}
}
}
JSON
```
For