Z-Image-Re-Turbo is a text-to-image generation model that has been optimized for de-reduction and re-acceleration based on the Z-Image-De-Turbo model. This model aims to balance the convenience during training and the speed during inference. It restores the fast generation ability close to the original Turbo model while maintaining the same training-friendly features as Z-Image-De-Turbo, enabling it to be perfectly compatible with a large number of trained LoRA models in the Z-Image ecosystem.

MultimodalDiffusers

D-ART - LoRA - Z-Image-Turbo is an image generation model based on the Tongyi-MAI/Z-Image-Turbo base model. It is a LoRA (Low-Rank Adaptation) adapter designed to inject a specific 'D-ART' artistic style into the generated images. This style is characterized by generating illustrations of confident and dynamic young female characters, especially emphasizing the curves and dynamic poses of the characters. Users can activate this style by using specific trigger words to generate diverse images with unified aesthetic features.

Computer VisionDiffusers

This is a LoRA model based on Qwen Image Edit 2509, specifically designed to convert input images into sketch comic artworks with exaggerated features. The model can create humorous and artistic comic images for human and animal subjects, highlighting facial features and characteristics.

Computer VisionDiffusersMultiple Languages

This is a Transformer model uploaded to the Hugging Face model hub. Specific functions and features require further information supplementation.

Natural Language ProcessingTransformers

Trouper-12B GGUF is a text generation model developed by DarwinAnim8or based on the original model. It is specifically optimized for role-playing and creative writing scenarios. This model is fine-tuned from Mistral-Nemo-Base-12B and trained using a custom 'Actors' dataset. The generated text is more natural and avoids the common 'AI language' features.

Natural Language ProcessingGgufEnglish

This is a lightweight sentence embedding model based on the MiniLM architecture, specifically designed to generate high-quality sentence vector representations. The model is trained using MultipleNegativesRankingLoss and performs excellently in sentence similarity calculation and feature extraction tasks.

Natural Language ProcessingSentence-transformers

Qwen-Image-Edit-Cosplay LoRA is an anime character dressing migration model fine-tuned based on the Qwen-Image-Edit-2509 image editing model. It focuses on accurately migrating the hairstyles, clothing, and decorative props of anime characters to real human images while maintaining the facial features of the original person.

Computer VisionDiffusers

MiniMax-M2 is a Mixture of Experts model designed for efficient coding and agent workflows. It has a total of 230 billion parameters and 10 billion active parameters. This model performs excellently in coding and agent tasks, and features low latency, low cost, and high throughput, effectively improving work efficiency.

Natural Language ProcessingTransformers

Lamapi/next-12b is a multilingual large language model with 12 billion parameters, offering multiple quantized versions and supporting various natural language processing tasks such as text generation, question answering, and chatting. This model is trained on multiple domain datasets and features high efficiency and lightweight characteristics.

Natural Language ProcessingTransformersMultiple Languages

Granite-4.0-H-1B-8bit is a small language model in the IBM Granite series, specifically optimized for Apple Silicon chips. It uses 8-bit quantization technology, has 1B parameters, and features efficient inference and low resource consumption.

Natural Language ProcessingMlx

This is a lightweight text summary generation model based on the Google T5 Efficient Tiny architecture. It is trained using the shorecode/summary-collection-200k-rows dataset and is specifically designed for automatic text summarization tasks. It features efficient inference and low resource consumption.

Natural Language ProcessingTransformersEnglish

This model uses the StyleGAN1 architecture and is trained with the StyleGAN2 regularization technique. It can generate realistic facial images with a resolution of 128×128. The model is trained on the FFHQ dataset and supports features such as style mixing, noise injection, and truncation tricks.

Computer VisionSafetensorsOther

This is a quantized version of the zfj1998/A2Search-3B-Instruct model, providing GGUF format files of various quantization types, which is convenient for users to choose and use according to their needs. This model mainly supports English tasks and has the feature of multiple quantization type selection.

Natural Language ProcessingTransformersEnglish

FIBO is the first open-source text-to-image model specifically trained for long structured descriptions, setting new standards for controllability, predictability, and feature decoupling. This model has 8 billion parameters and is trained using only licensed data, supporting the needs of professional workflows.

MultimodalDiffusersEnglish

This is a Vision Transformer model based on RoPE (Rotary Position Encoding) with a parameter scale of 150M and a patch size of 14x14. The model has been pre-trained on a diverse dataset of approximately 21 million images and can be used as a general visual feature extractor.

Computer VisionPytorch

This is a vision transformer model pre-trained on DINOv2, specifically optimized for retinal color fundus photos. The model is trained on a subset of the MEH AlzEye dataset and is provided in a format compatible with Hugging Face Transformers, facilitating retinal image feature extraction and downstream medical imaging tasks.

Computer VisionTransformersOther

Ming-UniAudio is an innovative unified speech framework that integrates speech understanding, generation, and editing functions into one. It uses a unified continuous speech tokenizer, which can effectively fuse semantic and acoustic features in an end-to-end model. It is the first system that can achieve general and free-form speech editing only through natural language instructions.

MultimodalSafetensors

F2LLM is an open-source text embedding model fine-tuned based on Qwen3-0.6B. Through single-stage training on 6 million high-quality query-document pairs, it achieves embedding performance comparable to the current state-of-the-art models. This model is specifically designed for feature extraction tasks and supports English text processing.

Natural Language ProcessingTransformersEnglish

This is a pre-trained model based on the Transformer architecture. The specific functions and features need to be supplemented according to the actual model information. The model supports multiple downstream tasks and has good generalization ability.

Natural Language ProcessingTransformers

This is a vision Transformer model based on the DINOv3 framework, trained on the LVD-1689M dataset from the DINOv3 ViT-7B model through knowledge distillation technology. This model is specifically designed for image feature encoding and can efficiently extract image feature representations, suitable for various computer vision tasks.

Computer VisionTimm