# Non-autoregressive model

Meissonic is a non-autoregressive masked image modeling text-to-image model capable of generating high-resolution images, specifically designed to run on consumer-grade GPUs.

This is a large Conformer-CTC model for Russian automatic speech recognition, trained on approximately 1,636 hours of Russian speech data with about 120 million parameters.

This is a large automatic speech recognition (ASR) model based on the Conformer architecture, supporting English speech transcription and trained using CTC loss function.