Rein r1 family decision model
Rein r1 uses open decision models to steer chat replies and tool calls. Compare its two sizes, rule-based interface and maker-reported test limits.
By jmc0 · Weights upload: 6 Nov 2026
Open 0.8B and 4B decision models that steer a separate chat model using rules supplied per request. Turn and partial-draft judgments can request tools, enforce an opening phrase or stop a reply. Rein returns decisions; the chat model writes the text.
Quick answers
What is Rein r1 family by jmc0?
Rein r1 family: Open 0.8B and 4B decision models that steer a separate chat model using rules supplied per request.
Who makes Rein r1 family?
Are Rein r1 family's weights open?
Rein r1 family: Public merged backbone and decision-head weights Licence: Apache-2.0, per both cards
Facts
- Variants
- Rein-r1-0.8B and Rein-r1-4B
- Inputs
- Rule sheets and text conversations; English and Portuguese best measured
- Runtime
- Custom Transformers decision and session-control code
- Licence
- Apache-2.0, per both cards
- Access
- Public merged backbone and decision-head weights
Claims and sources
- The maker reports 96.8% decision accuracy and 0.003 ECE for 4B on 28,430 questions about held-out rules. Labels are synthetic and audited by another model. End-to-end chat tests still contain rule violations; they are small sampled sessions judged by Qwen. vendor reportedsource
Links
Source: Model card and weights → Source: 0.8B card and weights →
Related projects
No related projects are listed in the registry.