This release contains the main Stage-0 direct scorers and Stage-1 reasoning-context checkpoints used in the Q-Scorer+ pipeline. The checkpoints are compact Q-Scorer+ checkpoints: they contain LoRA adapters, the score head, the score-token delta, tokenizer files, and reports. They are not merged 8B weight dumps. The base model is fetched from Qwen/Qwen3-VL-8B-Instruct. Clone this repository together with the repository root, or copy qscorerplus/ to your Python path. Then load a checkpoint with: The loader resolves the base model from Qwen/Qwen3-VL-8B-Instruct. No local base-model files are required as long as HuggingFace Hub is reachable. For offline use, download the base model first and…
Independent publisher
Jacuzzi
2kxx
Models in Library1
Datasets in Library0
Models on Hugging Face5
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
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.