Ru-Jev decision model
Ru-Jev is a Russian-first decision model built on mmBERT-base with 307M parameters. It runs on CPU and has a separate scoring head.
By besktech (beskvladimir-create) · First Hugging Face repository creation: 1 Nov 2026
A Russian-first typed-decision model built on mmBERT-base (307M), with a separate scoring head. It runs on CPU and supports yes/no, choice and ordinal scales. The maker says it is unaffiliated with TypeSafe AI and was not trained on Jev outputs. The v1.3 card documents improved option-wording robustness and withdraws an earlier calibration comparison.
Quick answers
What is Ru-Jev by besktech (beskvladimir-create)?
Ru-Jev: A Russian-first typed-decision model built on mmBERT-base (307M), with a separate scoring head.
Who makes Ru-Jev?
Ru-Jev: besktech (beskvladimir-create)
Facts
- Size
- 307M (mmBERT-base)
- Licence
- Code MIT; weights tagged MIT on Hugging Face
- Run it
- Load the backbone, head.safetensors and calibration.json using the Hugging Face card's Python example; the standard inference widget cannot load the separate head
- Release date
- First Hugging Face repository creation: 01 Nov 2026; v1.3 card checked 05 Nov 2026
- Calibration
- Optional label-specific additive biases can change the selected answer; temperature alone cannot
Claims and sources
- For v1.3, the maker reports 0.823 micro accuracy and 0.0185 ECE on 13,373 decisions, with temperature fitted on a disjoint half. Its 0.0159 ECE fitted and scored on the same data is in-sample. vendor reportedsource
- On six tasks with identical rows and option wordings, v1.1 and v1.3 each get 8,120 of 10,157 decisions right (0.7994). The card withdraws its claim of better calibration after temperature fitting: split variation reverses the comparison. Word-sense questions supply 84% of these rows. vendor reportedsource
- On the fifteen routing categories whose jargon labels do not coincide with training wordings, the maker reports accuracy rising from 0.624 for v1.1 to 0.838 for v1.3. Each recipe was run once; this is not a multi-seed result. vendor reportedsource
- The maker reports v1.3 accuracy 0.568 rising to 0.640 on news rubrics under label-specific bias fitting; v1.1 reaches 0.673 under the same fit. The fitted biases belong to that option set and do not transfer to another set. This is a calibration-tool update, without a new backbone release. vendor reportedsource
Links
Source: Weights and v1.3 model card → Source: GitHub → Source: Release metadata →
Related projects
No related projects are listed in the registry.