Organization
LuVox
luvox-ai
Models in Library2
Datasets in Library1
Models on Hugging Face13
Followers7
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, Speech recognition.
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
Datasets
Single val split: 1,013 retained records from 303 source train + 744 source validation records of tridm/VMLU. Source train: 292 retained / 11 excluded; source validation: 721 retained / 23 excluded. All 58 subjects remain represented. No test data was downloaded or included. Only these nine columns are exported in the Parquet dataset: Every promptaudio begins with: The remaining answer-letter instructions and all original choices are unchanged. For text evaluation use prompt. For audio evaluation supply promptaudio together with audio; do not send question, transcription or ttstext as additional model context. The audio contains the question only, without options, answer keys or model…