Jev directory
CommunityJev Models & Reimplementations
ROR Set Point Set Model mmBERT Base (OpenAlex Jev Distill)
A cross-encoder set model for ROR set point matching, distilled from OpenAlex's Jev judgments. Given an affiliation string and its candidate ROR records, it gives each candidate the probability that the string names it.
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
Highlights
- Cross-encoder set model fine-tuned from jhu-clsp/mmBERT-base.
- Scores each candidate ROR record with the probability that the affiliation string names it.
- Stage 1 learns from teacher judgments of openalex-affiliation-matcher v3.0.0, which come from Jev.
- The card documents a mention mask and reports evaluation with and without it.
- MIT licensed, 306M parameters in BF16 safetensors.
Questions about this entry
Short answers drawn from the linked project or page.
It reads an affiliation string together with its candidate ROR records and gives each candidate the probability that the string names it, learning from OpenAlex's Jev judgments.
Load it as a feature-extraction pipeline: from transformers import pipeline; pipe = pipeline("feature-extraction", model="adambuttrick/ror-set-point-set-model-mmbert-base-openalex-jev-distill"). The card also shows AutoTokenizer and AutoModel usage.
It is a fine-tune of jhu-clsp/mmBERT-base, a ModernBERT model, and is tagged as a cross-encoder for feature extraction.
Its stage 1 learns from the teacher judgments of openalex-affiliation-matcher v3.0.0. The card states those judgments come from Jev, a closed commercial decision model from TypeSafe AI.
The model card has a dedicated section on the mention mask and reports evaluation results both with and without it, alongside results on other test splits.
The matcher uses ror-set-point-set-model-mmbert-base-mask, which learns from the judgments of StartLux-Decision v1.1 27B Instant, a model with open weights. The card compares the two pipelines, and this repository is a copy of revision a43d230 of the mask variant.
The repository is MIT licensed. Its safetensors weights hold 306,939,648 parameters in BF16, roughly 613 MB, in a single unsharded file.
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 Jev-style decision model built on LiquidAI's LFM2.5-350M backbone, a clone of kev. Serves the System One API contract so the TypeSafe SDK works against a local server, returning probabilities for typed questions from one forward pass.
Jev Models & Reimplementations#open-models#python#local
Communityfranckverrot/lev
v-jev 1.4 is a Jev-style decision model on Hugging Face in 0.8B and 4B sizes, built on Qwen3.5 backbones. It returns probabilities for typed questions in one forward pass and adds value extraction and image decisions.
Jev Models & Reimplementations#open-models#classification#calibration
Community
A 1.7B-parameter SocialJev preview model on Hugging Face, fine-tuned from Qwen3-1.7B for typed classification in the social sciences. Runs through Transformers with trust_remote_code or the social-jev package with a vLLM backend, returning a choice and probabilities per query.
Jev Models & Reimplementations#open-models#classification#python
Community
A 4B-parameter SocialJev preview model on Hugging Face, fine-tuned from Qwen3-4B for typed classification in the social sciences. Loads through Transformers with trust_remote_code or the social-jev package with a vLLM backend, returning a choice and probabilities per query.
Jev Models & Reimplementations#open-models#classification#python
Community
A 380M-parameter UltraBERT encoder fine-tuned as a Jev-style decision model: it reads a state, a question, and a list of options and picks one option in a single forward pass. Apache-2.0 weights hosted on Hugging Face, scored by softmax over each option's " yes" logit.
Jev Models & Reimplementations#open-models#python#classification
Community
A Hugging Face checkpoint that fine-tunes UltraBERT-380M into a Jev-style decision model: it reads a state, a question and a list of options and picks one in a single forward pass. Reports 72.6% mean accuracy over 13 datasets on the test split.
Jev Models & Reimplementations#open-models#classification#local
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.