Jev directory
CommunityJev Models & Reimplementations
SocialJev 1.7B Preview
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.
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
Highlights
- Fine-tuned from Qwen/Qwen3-1.7B and tagged for classification and feature extraction.
- Ships 1,720,577,025 BF16 parameters in a single unsharded safetensors file of about 3.4 GB.
- Loads via AutoModel with trust_remote_code=True and custom class model.SocialJevModel.
- The social-jev package adds a vLLM backend via uv pip install "social-jev[vllm]".
- Queries return .choice and .probabilities; results are ordered by document, then query.
Questions about this entry
Short answers drawn from the linked project or page.
Install torch, transformers, and pydantic, then call AutoModel.from_pretrained("social-jev/social-jev-1.7B-preview", trust_remote_code=True, device_map="auto"). The repository uses custom code with the class model.SocialJevModel, so trust_remote_code must be enabled.
The repository tags list base_model:Qwen/Qwen3-1.7B and base_model:finetune:Qwen/Qwen3-1.7B, so the preview is a fine-tune of Qwen3-1.7B.
The safetensors metadata reports 1,720,577,025 BF16 parameters in a single unsharded file, with a total file size of roughly 3.4 GB.
Install the optional dependencies with uv pip install "social-jev[vllm]", then construct SocialJev.from_pretrained("social-jev/social-jev-1.7B-preview", backend="vllm") and submit requests through the same typed interface.
Each result exposes .choice and .probabilities, for example Transport with {'Transport': 0.98, 'Education': 0.01, 'Healthcare': 0.01}. Results are ordered first by document, then by query for that document.
No. The model card states that SocialJev is under active development and that interfaces may change between early releases, so pin versions if you depend on the typed API.
No. The model can be used directly through Transformers without installing the SocialJev package; the package is only needed for the shared typed interface and the optional vLLM backend.
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
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 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
MLX precision variants of Jev-Style-2B-Decision-v4.2, a 1.88B-parameter text-only decision model for Apple silicon. Ships bf16 and 8-bit weights with a custom runtime that returns a probability for every supplied option.
Jev Models & Reimplementations#open-models#local#python
Community
LoRA adapter for Qwen3.5-4B that scores medical multiple-choice options in a single forward pass and returns a short JSON object with the chosen answer key and per-option probabilities, without explanations or chain-of-thought.
Jev Models & Reimplementations#open-models#python#classification
Community
Full-parameter fine-tune of Qwen3.5-4B for medical multiple-choice scoring. Given a case and candidate options, it returns the chosen option key with per-option probabilities as short JSON, without explanations or chain-of-thought.
Jev Models & Reimplementations#open-models#classification#python
Community
Russian-first open typed-decision model on Hugging Face. Given a context and a set of candidate options it returns a calibrated probability and picks one, covering yes/no, choice, and ordinal scale decisions. Built on mmBERT-base with 307M parameters under an MIT license.
Jev Models & Reimplementations#open-models#calibration#classification
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.