StableDelight is an image processing model that reveals hidden textures by eliminating specular reflections.

Computer VisionDiffusers

Kaviel is a text classification model fine-tuned based on roberta-base, specifically designed to classify text into six major threat categories: bank fraud, terrorist attacks, life threats, online scams, information leakage, and daily conversations, helping threat intelligence platforms automatically detect and manage threats.

Natural Language ProcessingTransformersEnglish

A high-performance cross-encoder model optimized for Japanese text reranking tasks, featuring a 24-layer architecture with 1024 hidden units

Natural Language ProcessingSafetensorsJapanese

This is a Japanese Reranker (Cross-Encoder) model for text ranking tasks, featuring 24 layers and a hidden layer size of 1024.

Natural Language ProcessingSafetensorsJapanese

Caduceus-Ph is a DNA sequence modeling model based on the MambaDNA architecture, with a hidden dimension of 256 and a 16-layer structure.

Scientific ComputingTransformers

smol_llama-101M-GQA is a small decoder model with a total of 101M parameters. It adopts the GQA (Grouped Query Attention) architecture, having 6 hidden layers and a hidden layer size of 768. This model was pre-trained on a single GPU for 5 compute days, supports a context length of 1024, and focuses on providing efficient text generation capabilities.

Natural Language ProcessingTransformersEnglish

Hubert (Hidden-Unit BERT) is a speech representation learning model proposed by Facebook, which uses self-supervised learning to directly learn speech features from raw waveform signals.

Audio ProcessingJaxKorean

A Chinese RoBERTa medium model pre-trained on CLUECorpusSmall, with 8 layers of network and a 512-dimensional hidden layer, suitable for various Chinese NLP tasks.

Natural Language ProcessingJaxChinese

A Chinese RoBERTa model pre-trained on CLUECorpusSmall, with a parameter scale of 8 layers and 512 hidden units.

Natural Language ProcessingJaxChinese

A Chinese RoBERTa model pre-trained on CLUECorpusSmall, with a parameter scale of 8 layers and 512 hidden units, supporting masked language modeling tasks.

Natural Language ProcessingJaxChinese

This is a Chinese pre-trained language model based on the RoBERTa architecture, with a parameter scale of 8 layers and 512 hidden units, suitable for various Chinese natural language processing tasks.

Natural Language ProcessingJaxChinese

A lightweight Chinese GPT2 model pre-trained on CLUECorpusSmall, with 6 layers/768 hidden units, suitable for Chinese text generation tasks

Natural Language ProcessingJaxChinese

A Chinese RoBERTa model pretrained on CLUECorpusSmall, with 8 layers and 512 hidden units, suitable for various Chinese NLP tasks.

Natural Language ProcessingJaxChinese

A Chinese RoBERTa model pre-trained on CLUECorpusSmall, featuring 8 layers and 512-dimensional hidden layers, suitable for various Chinese NLP tasks.

Natural Language ProcessingJaxChinese

BERT-Medium is a lightweight version of the BERT model released by Google, featuring fewer network layers and hidden units, making it suitable for resource-constrained environments.

Natural Language ProcessingJax

A variant of the BERT model pre-trained using knowledge distillation technology, with a hidden layer dimension of 240 and 12 attention heads, suitable for masked language modeling tasks.

Natural Language ProcessingTransformers

Chinese RoBERTa model pre-trained on CLUECorpusSmall, featuring 8 layers and 512 hidden dimensions, suitable for various Chinese NLP tasks.

Natural Language ProcessingJaxChinese

A lightweight model based on the BERT architecture, pre-trained using knowledge distillation techniques, with a hidden layer dimension of 256 and 4 attention heads.

Natural Language ProcessingTransformers

Chinese RoBERTa model pretrained on CLUECorpusSmall with 8 layers and 512 hidden units, suitable for various Chinese NLP tasks

Natural Language ProcessingJaxChinese

Chinese pre-trained language model based on RoBERTa architecture, with hidden layer dimension of 512 and 8 Transformer layers

Natural Language ProcessingJaxChinese