{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T18:12:11.088Z","data":{"recordId":"wsa_zwjdvrdbpq2n","score":1,"confidence":"high","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://huggingface.co/FINAL-Bench/Darwin-27B-ZTC-v2","publisher":"Hugging Face","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-27T01:40:35.745Z","detectedPublishedAt":"2026-10-08T15:28:29.000Z","rights":"original"}],"claims":[{"id":"e596012e-a846-482c-8033-182ac3d86e3b","kind":"capability","statement":"The model returns a probability for every option in a typed question.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"**A zero-token decision engine from the Darwin family, second version.** Darwin-27B-ZTC-v2 reads a piece of state and a typed question (`noul` yes or no, `choice` one of N labels, `score` an ordered rubric) and returns a probability for every option.","rights":"original"}}]},{"id":"9d4369ef-309e-4ca9-befb-3facfca681ec","kind":"capability","statement":"It uses one forward pass per question and generates no tokens.","citations":[{"source":1,"supportProbability":0.99,"supports":true,"quote":{"text":"It uses one forward pass per question and generates no tokens.","rights":"original"}}]},{"id":"f695c05e-f318-4b4a-86bc-76b171f8c25b","kind":"capability","statement":"Version 2 adds control and workflow decisions, including games, drones, retrieval control, entity alignment, and customer and incident workflows.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"v2 adds control and workflow decisions (games, drones, retrieval control, entity alignment, customer and incident workflows) on top of [FINAL-Bench/Darwin-27B-ZTC](https://huggingface.co/FINAL-Bench/Darwin-27B-ZTC).","rights":"original"}}]},{"id":"54037bf6-1594-4504-9446-72066fe37ad1","kind":"availability","statement":"The checkpoint is available at FINAL-Bench/Darwin-27B-ZTC-v2 on Hugging Face.","citations":[{"source":1,"supportProbability":0.76,"supports":true,"quote":{"text":"All 137 benchmarks complete, measured with the S1MB evaluator and its `autojev` adapter at revision `a8f283a`, then validated and merged by the S1MB maintainer ([results](https://huggingface.co/datasets/hotchpotch/s1mb-result/tree/main/FINAL-Bench__Darwin-27B-ZTC-v2)).","rights":"original"}}]},{"id":"fc045ef0-5faa-436a-9165-488c5d737769","kind":"deployment","statement":"The release includes a server that exposes a POST /v1/systemone endpoint and an in-process engine.","citations":[{"source":1,"supportProbability":0.97,"supports":true,"quote":null}]},{"id":"1710b972-7bb1-451d-8442-6972efca962b","kind":"recommendation-relevant","statement":"The backbone weights are BF16 and about 54 GB.","citations":[{"source":1,"supportProbability":0.98,"supports":true,"quote":{"text":"The layout is the same as v1: backbone weights (BF16, about 54 GB), `readout.safetensors`, `decision_config.json`, tokenizer, and the inference code in `autojev/` (from [autojev](https://github.com/denis-pplx/autojev), MIT, with text-only backbone support added).","rights":"original"}}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}