This is the third version of the monolingual encoder-decoder language model released by the HPLT project, specifically designed for German. It is a text-to-text transformer model trained based on the denoising objective, following the NorT5 architecture settings, and using a tokenizer trained with German HPLT data.

Natural Language ProcessingPytorchMultiple Languages

ARPG is an innovative autoregressive image generation framework capable of achieving BERT-style masked modeling through a GPT-like causal architecture.

Computer VisionDiffusers

Open-Sora is an open-source, efficient text-to-video generation model designed to lower the barrier to high-quality video generation technology.

MultimodalSafetensors

The Hunyuan Reward Model includes two LoRA models, HPS and MPS, focusing on aesthetic performance and prompt adherence respectively, suitable for text-to-image generation tasks.

MultimodalDiffusers

Open-source medical large language model, achieving state-of-the-art performance in multiple medical tasks, developed based on Qwen2.5-72B

Natural Language ProcessingTransformersEnglish

Qwen2.5-Aloe-Beta-7B is an open-source large medical language model that achieves state-of-the-art performance in multiple medical tasks. It is fine-tuned based on the Qwen2.5-7B architecture, and the training data covers 1.8 billion tokens of diverse medical tasks.

Natural Language ProcessingTransformersEnglish

Llama3.1-Aloe-Beta-8B is an open-source large medical language model that achieves state-of-the-art performance in multiple medical tasks.

Natural Language ProcessingTransformersEnglish

AiM is an unconditional image generation model based on PyTorch, integrated and pushed to Hugging Face Hub via PytorchModelHubMixin.

Computer VisionSafetensors

HPC-Coder-v2-6.7b is a large language model fine-tuned on high-performance computing domain instruction datasets, focusing on typical HPC topics such as parallel computing, performance optimization, and accelerator porting.

Natural Language ProcessingTransformersMultiple Languages

Aloe is a new generation of medical large language model family, which achieves high competitiveness compared to all previous open-source models within its scale range and reaches state-of-the-art results through model fusion and advanced prompting strategies.

Natural Language ProcessingTransformersEnglish

A Thai monolingual BERT model released by the HPLT project, trained using the LTG-BERT architecture, suitable for Thai text processing tasks.

Natural Language ProcessingTransformersOther

A monolingual Slovak BERT model released by the HPLT project, trained on the LTG-BERT architecture, suitable for masked language modeling tasks

Natural Language ProcessingTransformersOther

HPT is a transformer model that aligns different entities into a shared latent space, focusing on the study of expansion behaviors in policy learning.

MultimodalTransformers

The Poro 34B Chat version is an instruction-following model fine-tuned based on Poro 34B, supporting bilingual interaction in Finnish and English, jointly developed by Silo AI, the TurkuNLP group, and HPLT.

Natural Language ProcessingTransformersMultiple Languages

This is a Transformer-based machine translation model from Arabic to English, trained exclusively on HPLT data.

Natural Language ProcessingTransformersMultiple Languages

Arabic-English machine translation model trained on OPUS and HPLT data, available in both Marian and Hugging Face formats.

Natural Language ProcessingTransformersMultiple Languages

VQVAE is a video generation model based on the VQ-VAE architecture, cloned from the VideoGPT project, aimed at converting the model to Hugging Face format for easier loading.

Computer VisionTransformers

NorMistral-7b-scratch is a large-scale language model specifically designed for Norwegian, developed in collaboration by the Language Technology Group at the University of Oslo, the HPLT project, the National Library of Norway, and the University of Turku. This model is based on the Mistral architecture, has 7 billion parameters, and was pre-trained from scratch on a Norwegian corpus of 260 billion subword tokens. It is an important part of the NORA.LLM series.

Natural Language ProcessingTransformersOther

HPSv2-hf is the Huggingface CLIPModel version of the HPSv2 model, designed to predict human preferences for AI-generated images.

MultimodalTransformers

An open-source bilingual Chinese-English large language model based on LLaMA-2, continuously pre-trained on approximately 8.5 billion tokens, supporting a context window of 4096 tokens.

Natural Language ProcessingTransformersMultiple Languages