This is a LoRA fine-tuned model based on Qwen-Image-Edit-2509, specifically designed to solve the problem of secondary illumination processing for images. It can remove the illumination effect of the original image and regenerate realistic lighting and shadows for the target image according to the illumination and hue of the reference image, achieving the migration and fusion of image styles.

Computer VisionDiffusers

CodeV is a 7-billion-parameter vision-language model fine-tuned based on Qwen2.5-VL-7B-Instruct. Through two-stage training of supervised fine-tuning (SFT) and reinforcement learning (RL) based on tool-aware policy optimization (TAPO), it aims to achieve reliable and interpretable visual reasoning. It represents visual tools as executable Python code and ensures the consistency between tool usage and question evidence through a reward mechanism, solving the problem of irrelevant tool invocation under high accuracy.

MultimodalTransformersMultiple Languages

The Pikachu Language Model is a unique experimental language project initiated by Clemylia. It is trained completely from scratch and is based solely on the sound corpus of 'Pikachu', demonstrating the ability of small language models to create a native language identity.

Natural Language ProcessingTransformers

Z-Image is a powerful and efficient image generation model with 6 billion parameters. It can effectively solve the problems in the fields of image generation in terms of efficiency, quality, and functional diversity, providing users with high-quality image generation and editing services.

Computer VisionDiffusersEnglish

Wan2.2 is a major upgrade version of the basic video model, focusing on integrating innovative technologies such as effective MoE architecture, efficient training strategies, and multimodal fusion into the video diffusion model, bringing more powerful and efficient solutions to the field of video generation.

MultimodalDiffusersMultiple Languages

SteadyDancer is a powerful animation framework based on the image-to-video paradigm, specifically designed to generate high-fidelity and temporally coherent human body animations. This framework effectively solves the identity drift problem in traditional methods through a robust first frame retention mechanism, performs excellently in visual quality and controllability, and significantly reduces the training resource requirements.

Computer VisionDiffusers

OpenMMReasoner is a completely transparent two-stage multimodal reasoning solution, trained using supervised fine-tuning (SFT) and reinforcement learning (RL). A cold-start dataset of 874K samples was constructed in the SFT stage, and the RL stage used 74K samples to further enhance its capabilities, showing excellent performance in multimodal reasoning benchmark tests.

MultimodalTransformers

OpenMMReasoner is a fully transparent two-stage multimodal reasoning solution that covers supervised fine-tuning (SFT) and reinforcement learning (RL). By carefully constructing high-quality datasets, this solution surpasses powerful baseline models in multiple multimodal reasoning benchmark tests, laying a solid empirical foundation for future large-scale multimodal reasoning research.

MultimodalTransformers

This is a mixed quantization version based on the Qwen3 VL 8B model, which uses MXFP4 quantization technology and mixed weights on the dense model. While maintaining accuracy close to Q8, this model achieves a smaller file size and higher inference speed, making it an excellent solution for balancing efficiency and accuracy.

MultimodalSafetensorsOther

Cyber Intelligence Brain is a large language model specifically fine-tuned for Capture the Flag (CTF) challenges and network security tasks. It is based on the Llama-3.1-8B-Instruct architecture and can assist in solving CTF problems, analyzing security vulnerabilities, conducting reverse engineering and cryptography challenges, and providing vulnerability exploitation reasoning while adhering to ethical guidelines.

Natural Language ProcessingSafetensorsMultiple Languages

Qwen2.5 is a new-generation large language model launched by Alibaba. It has made important optimizations in the chat template, permanently disabled the thinking state, solved problems related to the multi-round chat template, and provided a more convenient experience for experiments and research.

Natural Language ProcessingTransformers

A text generation model developed based on the MLX library, focusing on natural language processing tasks and providing developers with an efficient text generation solution.

Natural Language ProcessingMlxEnglish

MM-VAE Lyra is a multimodal variational autoencoder specifically designed for text embedding conversion, using geometric fusion technology. It combines the CLIP-L and T5-base models and can effectively handle the encoding and decoding tasks of text embeddings, providing an innovative solution for multimodal data processing.

TabPFN-2.5 is a tabular foundation model based on the Transformer architecture. It utilizes context learning technology and can solve tabular prediction problems in a single forward propagation, providing efficient regression and classification solutions for structured tabular data.

This is a quantized version model based on Qwen3-VL-30B, specifically designed for image-text to text conversion tasks. The model optimizes the original model through quantization technology, providing a more efficient inference solution.

MultimodalGguf

SDXL-Deepfake-Detector is a tool for accurately detecting AI-generated faces. It focuses on maintaining the authenticity of the digital world and provides a privacy-protected and open-source solution to combat visual misinformation. This model achieves lightweight and highly accurate detection through fine-tuning a pre-trained model.

Computer VisionSafetensors

Kimi Linear is a hybrid linear attention architecture that outperforms traditional full attention methods in various scenarios, including short and long contexts, as well as reinforcement learning extension mechanisms. It can effectively solve the problem of low efficiency of traditional attention mechanisms in long context tasks and bring more efficient solutions to fields such as natural language processing.

Natural Language ProcessingTransformers

This project is the result of quantizing the MiniMax-M2 model to 3 bits using the mlx-lm library through Dynamic Weight Quantization (DWQ). It can efficiently perform text generation tasks under resource-constrained conditions, providing a more lightweight solution for related applications.

Natural Language ProcessingMlx

Scout is a professional language model fine-tuned by VANTA Research based on the Google Gemma 3 4B Instruct architecture. It excels in constraint-aware reasoning and adaptive problem solving, and is擅长侦察式情报收集 should be translated as is good at scouting-style intelligence gathering。擅长侦察式情报收集、系统问题分解和自适应解决方案生成。It is good at scouting-style intelligence gathering, system problem decomposition, and adaptive solution generation.

Natural Language ProcessingTransformersEnglish

This is an image text recognition model based on DeepSeek-OCR, which specifically solves the compatibility issues in the latest version of the transformers library, enabling the model to run smoothly in the latest versions such as transformers v4.57.1.

MultimodalTransformersEnglish