# Fast Inference Optimization

BRIA 2.3 is a text-to-image model trained on legally compliant datasets, integrating technological innovation with ethical responsibility and legal safeguards, setting a new benchmark for the AI industry.

This is a distilled model based on the GPT2 architecture, specifically trained for typed story generation tasks with 84 million parameters.

A Telugu automatic speech recognition model fine-tuned based on OpenAI Whisper-large-v2, trained on multiple public Telugu datasets