# Zero-Shot Learning

Nalgotic Dreams is a text-to-image model based on Stable Diffusion XL, specializing in generating high-quality anime-style images, particularly bright and detailed illustrations of girl characters.

OpenVision is a fully open, cost-effective family of advanced vision encoders focused on multimodal learning.

OpenVision is a fully open, cost-effective advanced vision encoder family focused on multimodal learning.

OpenVision is a fully open, cost-effective advanced vision encoder family focused on multimodal learning.

GLM-4-32B-0414 is a large-scale language model developed by THUDM, with 32 billion parameters, suitable for various text generation tasks.

PURE is the first framework to employ a Multimodal Large Language Model (MLLM) as the backbone network for solving low-level vision tasks.

Qwen2.5-VL-72B-Instruct is a large-scale vision-language model developed by the Tongyi Qianwen team, supporting multimodal understanding and generation of images and text.

A vision Transformer model based on SigLIP 2, designed for image feature extraction and suitable for multilingual vision-language tasks.

Vision Transformer model based on SigLIP 2 architecture, pre-trained on WebLI dataset, with attention pooling head removed and global average pooling applied

A ViT image encoder based on SigLIP 2 for extracting image features, supporting multilingual vision-language tasks.

Phi-4-Model-Stock-v2 is a large language model merged from multiple Phi-4 variant models using the model_stock merging method, demonstrating strong performance across multiple benchmarks.

This model is a video language-guided reasoning segmentation model developed based on LLaVA-Phi-3-mini-4k-instruct, focusing on object segmentation tasks in videos.

OmniGen is a unified image generation model that supports multiple image generation tasks.

Lumina-mGPT is a family of multimodal autoregressive models, excelling in generating flexible and realistic images from text descriptions, and capable of performing various vision and language tasks.

PAVE is a model focused on repairing and adapting video large language models, aiming to enhance the conversion capability between video and text.

Med42-v2 is a clinically aligned large language model suite developed by M42, based on the LLaMA-3 architecture, available in 8B or 70B parameter versions, designed to provide high-quality medical Q&A capabilities.

Med42-v2 is an open-access clinical large language model suite developed by M42, built on LLaMA-3, containing 8 billion or 70 billion parameters, capable of high-quality medical question answering.

This is an AI model released under the Apache-2.0 license, with specific functionalities to be determined based on the actual model type

Libra is a decoupled vision system built upon large language models, possessing fundamental multimodal understanding capabilities.

Yuna AI is a virtual companion model designed for emotional connection, offering deep interactive experiences beyond traditional assistants.

ViTamin-XL-256px is a vision-language model based on the ViTamin architecture, designed for efficient visual feature extraction and multimodal tasks, supporting high-resolution image processing.

MoAI is a large-scale language and vision hybrid model capable of processing both image and text inputs to generate text outputs.

LLaVA is a large language and vision assistant model capable of handling multimodal tasks involving images and text.

EraseDraw is a diffusion model-based image editing tool capable of inserting or modifying objects in images based on text prompts.

supermario-slerp-v2 is a text generation model created by merging two 7B-parameter models using the SLERP method, demonstrating outstanding performance across multiple benchmarks.

This is an image captioning model based on the Vision Encoder-Decoder architecture, capable of generating natural language descriptions for input images.

An instruction-tuned version based on Stable Diffusion v1.5, specifically designed for image cartoonization

This model is used to predict the relevance of query-document pairs, suitable for information retrieval tasks.

The base version of Chinese CLIP, using ViT-B/16 as the image encoder and RoBERTa-wwm-base as the text encoder, trained on a large-scale dataset of approximately 200 million Chinese image-text pairs.

This is an image classification model generated by HuggingPics, specifically designed to recognize specific categories of images.

This is an image classification model generated by HuggingPics, capable of accurately identifying animals such as horses, llamas, and zebras.