# Multi-dataset training

This project provides a static quantized version of the richinfoai/ritrieve_zh_v1 model. Through quantization, it reduces storage space and computational resource requirements while maintaining certain performance.

A large-scale Vietnamese automatic speech recognition model based on the ChunkFormer architecture, fine-tuned on approximately 3000 hours of publicly available Vietnamese speech data, with excellent performance.

This is a fine-tuned model based on BERT, used to classify user inputs into 82 different intents, suitable for dialogue systems and natural language understanding tasks.

Aura-4B is a quantized version based on AuraIndustries/Aura-4B, using llama.cpp for imatrix quantization, supporting multiple quantization types, suitable for text generation tasks.

Whisper-medium model optimized for Vietnamese speech recognition tasks, fine-tuned on 1308 hours of Vietnamese data

A Whisper variant model focused on speech recognition in the Japanese anime field, optimized for the characteristics of anime audio

Llama3-Aloe-8B-Alpha is an 8B parameter large language model focused on the fields of biology and medicine, offering a quantized version in GGUF format

An anime-style text-to-image generation model based on SDXL, suitable for beginners, capable of generating high-quality anime characters and stylized images.

BiRefNet is a high-resolution binary image segmentation model based on bilateral reference, focusing on background removal and mask generation tasks.

A bilateral reference framework for high-resolution binary image segmentation, focusing on background removal and mask generation tasks

Octo is a multimodal foundation model for robotics, capable of predicting robot actions through visual and language inputs.

Vision Transformer model trained with self-supervised DINOv2, specifically designed for encoding chest X-ray images

Vietnamese automatic speech recognition (ASR) model fine-tuned based on OpenAI Whisper-tiny architecture, demonstrating excellent performance on multiple Vietnamese datasets

Finance LLM is a language model specialized in the financial domain, based on the Llama architecture, fine-tuned with datasets such as OpenOrca, Lima, and WizardLM.

This model is a zero-shot classification model based on the DeBERTa-v3-large architecture, primarily trained on five NLI datasets, suitable for tasks that strictly follow the original NLI task format.

The Silver Retriever model encodes Polish sentences or paragraphs into a 768-dimensional dense vector space, suitable for tasks like document retrieval or semantic search.

This model takes raw audio signals as input and outputs age predictions and gender probabilities (child/female/male), along with the pooled state of the last transformer layer.

Silver Retriever is a neural retrieval model specifically designed for Polish language, focusing on sentence similarity and paragraph retrieval tasks.

This is a version of thennal/whisper-medium-ml converted to the CTranslate2 model format for Malayalam speech recognition

This model is a Japanese speech recognition model fine-tuned based on openai/whisper-small, supporting Japanese speech-to-text tasks.

This reward model is trained to predict which generated answer human evaluators would prefer for a given question.

This is a Korean text summarization model based on the T5 architecture, specifically designed for Korean text summarization tasks. It is trained on multiple Korean datasets by fine-tuning the paust/pko-t5-base model.

French speech recognition model fine-tuned from openai/whisper-large-v2, trained on over 2200 hours of French audio data

A Mongolian speech recognition model fine-tuned on Mongolian datasets based on OpenAI's whisper-large-v2 model, supporting automatic speech recognition tasks in Mongolian.

This is a natural language inference (NLI) model based on the T5-XXL architecture, used to predict entailment relationships between text pairs ('1' for entailment, '0' for non-entailment).

A Portuguese semantic text similarity model fine-tuned based on the BERTimbau large model, capable of mapping sentences to a 1024-dimensional vector space

This is an Automatic Speech Recognition (ASR) model developed by NVIDIA, based on the Conformer-Transducer architecture, with approximately 600 million parameters, specifically designed for English speech transcription.

This model is a Thai automatic speech recognition model trained on the CommonVoice V8 dataset, incorporating the DeepCut tokenizer and language model to improve recognition accuracy.

NLI model fine-tuned on DeBERTa-v3-large, achieving state-of-the-art performance on multiple NLI datasets

Vietnamese speech recognition model based on Wav2vec2, fine-tuned on 160 hours of Vietnamese speech data

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

This is a Wav2vec 2.0 model fine-tuned for Portuguese, trained on multiple Portuguese speech datasets including CORAA, CETUC, MLS, etc.

This is a Wav2vec 2.0 model fine-tuned for Portuguese, trained using multiple Portuguese speech datasets including CORAA, CETUC, and Multilingual LibriSpeech.

A large-scale English automatic speech recognition model based on the Wav2Vec2 architecture, fine-tuned on 4500 hours of multi-source speech data, supporting decoding with a language model

This is a Dutch automatic speech recognition (ASR) model fine-tuned based on the XLS-R 1 billion parameter model, trained on multiple datasets including Common Voice 8.0, supporting 16kHz sampling rate audio input.

This is an Italian automatic speech recognition model based on the XLS-R 1B architecture, fine-tuned on multiple Italian datasets

Swahili automatic speech recognition model fine-tuned on XLSR-53 large model, supports 16kHz sampling rate audio input

This is a Polish automatic speech recognition (ASR) model fine-tuned based on the XLS-R 1-billion parameter model, trained on datasets such as Common Voice 8.0, supporting 16kHz sampling rate audio input.

This is a sentence embedding model based on the MPNet architecture, capable of mapping text to a 768-dimensional vector space, suitable for semantic search and sentence similarity tasks.

A Chinese extractive QA model based on the RoBERTa architecture, suitable for tasks that extract answers from given texts.