This project provides a quantized version (Int4) of Qwen/Qwen3-VL-32B-Thinking, aiming to lower the hardware threshold for model deployment and make the knowledge of advanced visual language models accessible to everyone.

MultimodalGguf

LongCat Image Editing is the image editing version of Longcat-Image. It is an open-source image editing model that supports both Chinese and English. It has reached the advanced level of open-source models in terms of instruction following, image quality, and visual consistency, and has excellent precise editing and consistency preservation capabilities.

MultimodalTransformersMultiple Languages

Rnj-1-instruct is an 8-billion-parameter open-weight dense model trained from scratch by Essential AI. It is optimized for code, STEM fields, agent capabilities, and tool invocation. It performs excellently in multiple programming languages and also demonstrates strong capabilities in mathematics and science, comparable to the most advanced open-weight models.

Natural Language ProcessingTransformers

NewBie image Exp0.1 is an efficient basic image generation model developed based on the Next-DiT architecture, specifically designed for generating high-quality anime-style images. It integrates advanced text encoders and visual components, supports natural language and structured tag input, and is a powerful tool for multi-character anime image generation.

Computer VisionDiffusersEnglish

TomoroAI/tomoro-colqwen3-embed-4b is an advanced ColPali-style multimodal embedding model that can map text queries, visual documents (such as images, PDFs) or short videos into aligned multi-vector embeddings. This model combines the advantages of Qwen3-VL-4B-Instruct and Qwen3-Embedding-4B, performs excellently in the ViDoRe benchmark test, and significantly reduces the embedding space occupation.

MultimodalTransformersOther

Fara-7B is an efficient small language model specially designed by Microsoft for computer usage scenarios. It has only 7 billion parameters and performs excellently in advanced user tasks such as web operations, competing with larger agent systems.

MultimodalGguf

Nanbeige 4-3B-Thinking-2511 is the latest enhanced version of the Nanbeige series. Through advanced distillation technology and reinforcement learning optimization, it achieves powerful reasoning ability within a compact 3B parameter scale. This model has achieved state-of-the-art (SOTA) results among models with less than 32B parameters in benchmark tests such as Arena-Hard-V2 and BFCL-V4.

Natural Language ProcessingTransformersMultiple Languages

GLM-4.5-Iceblink-v2-106B-A12B-FP8 is a version based on the GLM-4.5-Iceblink-v2-106B-A12B model, using the most advanced mixture of experts quantization method for FP8 quantization. This model is specifically optimized for Ada, Hopper, or Blackwell series GPUs that support hardware FP8, significantly improving inference efficiency while maintaining high-quality output.

Natural Language ProcessingSafetensors

VieNeu-TTS-1000h is an advanced Vietnamese edge-side text-to-speech model. It is trained on approximately 1000 hours of high-quality Vietnamese speech data, has an instant voice cloning function, supports seamless switching between Vietnamese and English, and can synthesize 24kHz waveforms in real-time on CPU or GPU.

Audio ProcessingTensorboardOther

NVIDIA Nemotron Parse v1.1 TC is an advanced document semantic understanding model that can extract text and table elements with spatial positioning from images and generate structured annotations, including formatted text, bounding boxes, and semantic categories. Compared with the previous version, the speed is increased by 20%, and the page order of unordered elements is retained.

MultimodalTransformers

NVIDIA Nemotron Parse v1.1 is an advanced document parsing model specifically designed to understand document semantics and extract text and table elements with spatial positioning. It can convert unstructured documents into machine-readable structured representations, overcoming the limitations of traditional OCR in handling complex document layouts.

MultimodalTransformers

Rnj-1 is a series of 8 billion parameter, open-weight dense models trained from scratch by Essential AI. This model is optimized for the code and STEM fields, excelling in programming, mathematical reasoning, agent tasks, and tool invocation, with capabilities comparable to the most advanced open-weight models.

Natural Language ProcessingTransformers

Parrotlet-e is an advanced multilingual medical embedding model optimized specifically for medical terms in various Indian languages. It is fine - tuned based on BAAI/bge - m3 and trained on over 18 million pairs of multilingual medical terms. It supports 12 Indian languages and English and has strong robustness against abbreviations, spelling variations, and colloquial expressions in clinical documents.

Natural Language ProcessingTransformers

The top model in the aquif-3.5 series, with advanced inference capabilities and a large context window of 1 million tokens, and has excellent performance in multiple benchmark tests. The comprehensive AAII score reaches 60 points.

Natural Language ProcessingTransformersMultiple Languages

The aquif-3.5 series is the masterpiece released on November 3, 2025, including two versions, Plus and Max, which provide advanced inference capabilities and an unprecedented 1 million token context window, achieving state-of-the-art performance in their respective categories.

Natural Language ProcessingTransformersMultiple Languages

Aquif-3.5-Max-42B-A3B is a large language model with 42 billion parameters, which has undergone MXFP4_MOE quantization processing. It optimizes inference efficiency while maintaining high-quality text generation capabilities. This model is based on an advanced mixture-of-experts architecture and is suitable for various natural language processing tasks.

Natural Language ProcessingGguf

WavJEPA is an audio foundation model based on the waveform joint embedding prediction architecture. It uses advanced semantic representation learning to address the deficiencies in speech unit or token-level representation learning. It significantly outperforms the state-of-the-art time-domain audio foundation models in many downstream benchmark tasks while requiring significantly less computational resources.

Audio ProcessingTransformers

This is a character detection model specifically optimized for Japanese video game text. It adopts the Pareto optimal design and achieves the best balance between accuracy and latency. The model redefines text recognition as a character detection task and is trained on Japanese video game data, achieving advanced recognition performance.

Computer VisionOnnxMultiple Languages

Next 12B is a multimodal vision-language model with 12 billion parameters based on Gemma 3. It is the most advanced open-source vision-language model in Turkey. This model performs excellently in text and image understanding, with advanced inference and context-aware multimodal output capabilities. It specifically provides professional-level Turkish language support and also has extensive multilingual capabilities.

MultimodalTransformersMultiple Languages

This is the MXFP4 quantized version of Qwen3-Next-80B-A3B-Thinking, which uses advanced quantization technology to significantly reduce storage and computational requirements while maintaining model performance. As an experimental version, it demonstrates the latest achievements in quantization technology.

Natural Language ProcessingGguf