Jev directory
CommunityJev Models & Reimplementations
ruvector-typesafe Banking77 Bank
A trained example bank for @ruvector/typesafe holding 9,993 labelled Banking77 utterances across 77 banking intents with frozen split assignments. Heads and temperature calibration are refit from the bank on load; no weights are stored.
- Reading time
- 2 min
- Published by
- Community
- Added
- Nov 2026
Highlights
- Stores 9,993 labelled examples across 77 fine-grained banking intents, with frozen split tags rather than weights.
- Held-out test split of 3,077 utterances: 87.0% accuracy with bge-small-en-v1.5 and 76.7% with all-MiniLM-L6-v2.
- Reported p50 latency is 6 ms (bge-small) and 15 ms (all-MiniLM-L6-v2); p95 is 10 ms and 16 ms.
- The bank is encoder-independent: both bundled encoders exported byte-identical banks for this dataset.
- The first decision after loading is slow; at probeIterations 4000 the head fit takes minutes, not milliseconds.
Questions about this entry
Short answers drawn from the linked project or page.
Install the library with npm install @ruvector/typesafe, download bank.json and questions.json from the model repo, then run npx typesafe decide with --questions, --bank and --embedder onnx.
A bank stores examples and frozen splits, never hyperparameters, and the head is refit from it on load. Omitting the options refits with library defaults of 400 iterations and head auto, which is a materially worse model than the one measured.
On the held-out test split of 3,077 utterances and 77 classes, bge-small-en-v1.5 reaches 87.0% and all-MiniLM-L6-v2 reaches 76.7%. Accuracy is reported on this dataset's own test split, not on your traffic.
No. It holds labelled examples with content-hashed split tags. The nearest-prototype or multinomial probe head and the temperature calibration are refit from the bank when the engine loads it, and a reloaded bank reproduces the trained engine's answers exactly.
No. It holds text and split tags, nothing encoder-derived, and both bundled encoders exported byte-identical banks for this dataset. Only the reported accuracy figures are encoder-specific.
It is English only, since both bundled encoders are English sentence encoders, and the label set is closed, so new intents need new examples and a refit. Loading a bank per request is also discouraged.
The @ruvector/typesafe code is MIT. The Banking77 dataset is CC-BY-4.0, and this bank redistributes the utterance text under that licence.
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
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 7.7M-parameter decision model on Hugging Face that reads text or speech and returns one probability per option for choice, score, and noul questions. Speech runs through femto-asr, giving a 10.1M-parameter spoken pipeline.
Jev Models & Reimplementations#open-models#classification#local
Community
Fine-tuned UltraBERT-380M decision model on Hugging Face that scores a state, question, and option list in one forward pass. Uses sigmoid of yes minus no normalized over options; 72.8% dev and 71.9% test accuracy across 13 datasets, with an evidence head documented on the card.
Jev Models & Reimplementations#open-models#arxiv#calibration
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
Fine-tuned UltraBERT-380M decision model on Hugging Face that reads a state, question, and option list and picks one option in a single forward pass. Scores options with a square-root softplus of yes minus no, normalized over options; 73.2% dev and 71.9% test accuracy over 13 datasets.
Jev Models & Reimplementations#open-models#arxiv#calibration
Community
A 46M-parameter Russian and English decision model published on Hugging Face, tagged for classification, NLI, toxicity, moderation, and calibrated distillation. It ships a custom NQ4ForDecision class and a chat template with an optional thinking block.
Jev Models & Reimplementations#open-models#classification#calibration
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.