# Encoder-decoder architecture

T0++ is a natural language processing model based on the T5 architecture, achieving zero-shot task generalization through multi-task prompt training, outperforming GPT-3 on various NLP tasks while being more compact.

TURNA is a Turkish encoder-decoder language model focused on enhancing understanding and generation capabilities.

A multilingual zero-shot classification model based on mT5-base, supporting bidirectional text classification in nearly 100 languages.

Pile-T5 Large is an encoder-decoder model trained on The Pile dataset based on the T5x library, primarily used for English text-to-text generation tasks.

State-of-the-art lightweight pretrained model for Vietnamese based on Transformer encoder-decoder architecture, specialized for text summarization tasks.

Advanced pre-trained encoder-decoder model for Vietnamese based on Transformer architecture

ruT5-base is a Russian text-to-text generation model developed by SberDevices, based on the T5 architecture with 222 million parameters and trained on 300GB of data.

T0pp is an 11-billion-parameter encoder-decoder model based on the T5 architecture, excelling in zero-shot task generalization with English natural language prompts, outperforming GPT-3 while being more compact.

This is an encoder-decoder model based on the RoBERTa architecture, specifically fine-tuned for sentence splitting tasks.

T0++ is an encoder-decoder model that demonstrates zero-shot task generalization capabilities on English natural language prompts, outperforming GPT-3 on many tasks while being 16 times smaller in size.