Qev fine-tunes Qwen into a decision model. Give it context, a question, and answer options; receive a decision and a probability for each option. One model supports Choice, Noul (yes/no), and Score (ordered ratings). Qev-9B combines a general decision-training dataset with additional alignment and rule-compliance examples, repeated three times during the second half of training. It uses the full candidate-interaction architecture shown below. Use Python 3.12 and install a hardware-compatible PyTorch 2.8.0 build, then install Qev from its source repository: The Qev loader downloads this checkpoint and the pinned Qwen base separately. Use the Qev Python or JSONL interface to load the adapter…
Independent publisher
Austin Scamander
AustinFu
Models in Library2
Datasets in Library1
Models on Hugging Face3
Followers—
Compare the published evidence
This profile connects the 3 model and dataset records shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Question answering.
Reported licenses: apache-2.0, other. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.
Models
Qev-2B is the compact member of the Qev decision-model family. It learns option probabilities and representation responses from Qev-9B, and supports Choice, Noul and Score with the same request format. Install Qev from the source repository using Python 3.12 and a hardware-compatible PyTorch 2.8.0 build. The loader downloads this adaptation package and the compatible Qwen base separately. The teacher is not needed for inference. Use revision="v0.1.0" to select the first published release. To fine-tune on your own data, use configs/qev-2b-finetune.json with python -m qev.train --init-checkpoint AustinFu/Qev-2B. To train with a teacher, follow the distillation guide or 中文指南. Accuracy (%).…
Datasets
1,842 synthetic training examples for Qev decision models. Each example provides a context, a question, explicit answer options and a reviewed hard label. This release collects the self-generated portion of the training recipe used for the released Qev-9B. The same source inputs were available in Qev-2B's initial teacher-probability training. It includes original hard labels, not teacher probability caches, response targets or the later distillation pairs. It is a subset of the original mixed training corpus. All examples are in the train split. There is no held-out evaluation split: these inputs have already been used in model development and training. Original wording, candidate order…