{"id":"zuse","slug":"zuse","name":"Zuse","orgId":"bineric","family":"zuse","releaseDate":null,"availability":"ga","openWeights":false,"license":null,"paramsTotal":27000000000,"paramsSource":"slug","contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://arxiv.org/abs/2611.06238v1","hfRepo":null,"modalities":["text"],"aliases":["Zuse","zuse","bineric zuse","zuse-4b","zuse-9b","zuse-27b","zuse 4b","zuse 9b","zuse 27b"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-07","announced":"2026-11-04","blurb":"Bineric AI's decision-model family from Norway: 4B, 9B and 27B research models return probabilities over caller-supplied options instead of generating text. A LoRA adapter and pointer head are trained on frozen Qwen bases, targeting 27 European languages. Zuse 27B has a public no-key live demo; 9B and 27B are offered through the API and commercial deployment options. Weights are not openly released, and Zuse 4B is not offered.","api":"live","openWeights":"no","params":{"text":"4B / 9B / 27B family","source":"Bineric Zuse paper, Table 1"},"facts":[{"label":"Role","value":"Text-only decision models: fixed options, yes/no and ordered score questions with probabilities; no free-text generation or image input"},{"label":"Bases","value":"4B: Qwen3.5-4B-Base; 9B: Qwen3.5-9B-Base; 27B: Qwen3.8-27B. Frozen bases with trained LoRA adapters and pointer heads; uses Kev serving stack, not a model router"},{"label":"Access","value":"Zuse 27B public live demo needs no API key. Bineric API offers 9B and 27B; 4B is described in the paper but not offered"},{"label":"Weights and license","value":"Zuse weights are not openly released; commercial API and deployment access. Apache-2.0 applies to the disclosed Qwen bases, not an open license for Zuse weights"},{"label":"Languages","value":"Trained targeting 27 European languages; broader language coverage is a lab evaluation claim, not 115 training languages"},{"label":"Context","value":"API lists 65,536 tokens per document; paper says accuracy above 32K is unmeasured and 24 GB-card deployments are limited to 8,192 tokens"},{"label":"Limitations","value":"Paper discloses development-set selection bias, single-seed results, undisclosed training data/recipe, task-dependent calibration, residual option-order sensitivity and prompt-injection susceptibility. Site cautions against using Zuse alone for employment, credit, benefits or legal-rights decisions"},{"label":"Related","value":"Separate from Qwen bases, TypeSafe Jev, Kev and Cloudflare Clef; 4B/9B/27B are one decision-model family"}],"claims":["Lab claim: Zuse 27B reaches 87.5% accuracy on 20,044 questions across 27 languages, with pooled expected calibration error 0.016 versus per-source weighted 0.035. This suite guided model selection; the paper does not claim an independent held-out evaluation.","Lab claim: on reading/topic evaluations across 210 languages, Zuse 27B reaches at least 85% in 115 languages. Training targets 27 European languages and wider transfer comes from the base model."],"sources":[{"kind":"lab","title":"Bineric Zuse paper v1 and dated submission metadata","url":"https://arxiv.org/abs/2611.06238v1","date":"2026-11-04"},{"kind":"lab","title":"Bineric Zuse paper: architecture, availability and limitations","url":"https://arxiv.org/html/2611.06238v1"},{"kind":"lab","title":"Bineric Zuse product and public live demo","url":"https://bineric.com/zuse"},{"kind":"lab","title":"Bineric founder announces Zuse paper","url":"https://x.com/Umair_/status/2118744647845417294","date":"2026-11-06"}],"firstBenchmark":null},"organization":{"id":"bineric","name":"Bineric AI","slug":"bineric","country":null,"url":"https://bineric.com/zuse","color":"#11110F","mark":"B"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Nov 4, 2026; no independent results yet.","lead":"As of Nov 22, 2026, Zuse is pending benchmarks on isitsota: announced Nov 4, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}