Article
CommunityJev Models & Reimplementations
RetiJev
A family of 0.8B-9B ophthalmic decision models that score candidate answers without autoregressive decoding, trained on RetiJev-Data: 78,158 clinical-decision and diagnosis questions and 50,074 guideline-grading questions. RetiJev-9B answers a case in 158 ms on one GPU.
- Format
- Article
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- RetiJev-Data contains 78,158 clinical-decision and diagnosis questions derived from PubMed Central articles.
- The corpus covers 531 eye diseases via MeSH vocabulary and 50,074 guideline-grading questions across 11 grading systems.
- RetiJev models range from 0.8B to 9B parameters and score candidate answers without autoregressive decoding.
- RetiJev-9B answers a case in 158 ms on one GPU, 3.5x faster than LLM generation and 470x faster than reasoning.
- RetiJev outperforms a 27B LLM and Jev 1.13 on guideline grading by 10.4-23.0 points across all model sizes.
Questions about this entry
Short answers drawn from the linked project or page.
RetiJev is a family of 0.8B-9B ophthalmic decision models that directly score candidate answers without autoregressive decoding, built for clinical decision support and guideline grading.
It was trained on RetiJev-Data, a source-grounded ophthalmology corpus with 78,158 clinical-decision and diagnosis questions from PubMed Central articles and 50,074 guideline-grading questions spanning 11 grading systems.
RetiJev-9B answers a case in 158 ms on one GPU, which is 3.5x faster than LLM generation and 470x faster than reasoning.
Across all model sizes, RetiJev outperforms both a 27B LLM and the commercial Jev 1.13 decision model on guideline grading by 10.4-23.0 points. RetiJev-9B also leads on clinical decisions by 5.4-6.5 points.
The models are released in the RetiJev repository on GitHub under the account Awenbocc.
RetiJev-Data includes a 24,662-question evaluation benchmark covering ophthalmic clinical decisions and guideline grading.
RetiJev is designed for decisions among predefined ophthalmic actions and guideline grading; it is not a general-purpose generative model for open-ended text or chat.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Standard One SH (StandardThinking)
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsMore Jev resources like this
Open-weight vision-language judge that takes a task instruction plus 10 frames of a robot manipulation attempt and predicts success or failure with calibrated probabilities in a single forward pass. Released with inference code and a 20,628-prediction evaluation set.
Jev Models & ReimplementationsArticle#arxiv#open-models#evaluation
Community
Hugging Face dataset of Jev-shaped retrieval and classification decisions built from MM-BRIGHT, holding Qwen3.6-35B-A3B-FP8 state, action, and hard-negative embeddings for post-training CLM heads.
Jev Models & Reimplementations#open-models#benchmark#evaluation
Community
Hugging Face dataset that renders HoVer claim-verification items as Jev Decision Index retrieval and classification requests, with Qwen3.6-35B-A3B-FP8 state, action, and hard-negative embeddings for post-training CLM heads.
Jev Models & Reimplementations#classification#benchmark#open-models
Community
arXiv paper testing whether the readout and adaptation choices used for direct selection still hold up in cross-request entailment reranking for JEV decision models. It combines a mean-span readout with selective backbone adaptation and reports accuracy on MC11, an average over eleven reranking configurations.
Jev Models & ReimplementationsArticle#arxiv#research#classification
Community
Apache-2.0 Hugging Face repository for RSI-JEV-27b, a 27B multimodal decision model based on Qwen3.8-27B that answers supplied multiple-choice questions over images, returning a selected option and probabilities for all options in one forward pass.
Jev Models & Reimplementations#open-models#benchmark#calibration
Community
arXiv paper proposing JevRec, which encodes recommendation items as variable-length semantic identifiers and adds a Jev-like decision modeling component with separate heads for length prediction and content generation, evaluated on three public datasets.
Jev Models & ReimplementationsArticle#arxiv#research#benchmark
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.