WavJEPA is an audio foundation model based on the waveform joint embedding prediction architecture. It uses advanced semantic representation learning to address the deficiencies in speech unit or token-level representation learning. It significantly outperforms the state-of-the-art time-domain audio foundation models in many downstream benchmark tasks while requiring significantly less computational resources.

Audio ProcessingTransformers

This is a quantized version of the deepcogito's cogito-v2-preview-llama-70B model, using the imatrix quantization technology of Llama.cpp. It significantly reduces the model storage and computational resource requirements while maintaining good performance, and is suitable for environments with limited resources.

Natural Language ProcessingGguf

This is a quantized version of the Austral-70B-Winton model by Delta-Vector. Through quantization technology, it reduces the storage and computational resource requirements of the model while maintaining good performance, making it suitable for scenarios with limited resources.

Natural Language ProcessingGgufEnglish

The AWQ 4-bit quantization version of QwQ-32B-Preview significantly reduces memory usage and computational requirements, making it suitable for hardware deployment with limited resources.

Natural Language ProcessingTransformersEnglish

LingLong is a lightweight Chinese pre-trained language model based on the GPT-3 architecture, with a parameter scale of 317 million, suitable for researchers with limited computational resources.

Natural Language ProcessingTransformersMultiple Languages

TinyLlama is a compact language model with 1.1 billion parameters, adopting the same architecture and tokenizer as Llama 2, suitable for applications with limited computational and memory resources.

Natural Language ProcessingTransformersEnglish

KuBERT is a Central Kurdish language model based on the BERT framework, designed to address the scarcity of Kurdish language resources and enhance computational linguistics capabilities.

Natural Language ProcessingTransformers

TinyLlama is a compact 1.1B-parameter language model pre-trained on 3 trillion tokens, suitable for applications with limited computational resources.

Natural Language ProcessingTransformersEnglish

BERT-Mini is a small BERT model developed by Google Research, fine-tuned on the SQuAD 2.0 dataset for question-answering tasks, suitable for environments with limited computational resources.

Natural Language ProcessingJaxEnglish

ElasticBERT is an efficient multi-exit BERT model that supports dynamic adjustment of computational resources.

Natural Language ProcessingTransformersEnglish