This is the first - generation powerful vibe - code large language model, optimized to directly generate natural language and code completion based on loosely structured 'atmosphere coding' prompts. Compared with early models, it has lower prompt engineering overhead, smoother latent space interpolation, and is easier to guide the generation of usable code.

Natural Language ProcessingTransformersEnglish

UserLM-8b-Q8_0-GGUF is a GGUF format version converted from the Microsoft UserLM-8b model, a large language model specifically optimized for user conversation scenarios. This model uses 8-bit quantization technology, significantly reducing the model size while maintaining performance, making it easy to deploy and run on various hardware.

Natural Language ProcessingGgufEnglish

NVIDIA GPT-OSS-120B Eagle3 is an optimized version based on the OpenAI gpt-oss-120b model. It adopts the Mixture of Experts (MoE) architecture, with a total of 120 billion parameters and 5 billion active parameters. This model supports both commercial and non-commercial use and is suitable for text generation tasks, especially for the development of AI Agent systems, chatbots, and other applications.

Natural Language ProcessingSafetensors

RND1 is an experimental diffusion language model with 30 billion parameters, adopting a sparse mixture of experts architecture. This model is converted from a pre-trained autoregressive base model, supporting diffusion-based text generation. Only 3 billion parameters are activated for each token, achieving a balance between computational efficiency and model capacity.

Natural Language ProcessingTransformers

The Earth & Dusk project is a text-to-image conversion project based on the OnomaAIResearch/Illustrious-xl-early-release-v0 base model, which uses the diffusers library to generate images. This project focuses on art creation and is licensed under the CREATIVE ML OPEN RAIL M license, supervised by 0FTH3N1GHT PRODUCTIONS.

Computer VisionDiffusersMultiple Languages

This is a speculative model based on the EAGLE-3 speculative decoding algorithm, specifically designed to work with the Qwen/Qwen3-32B model. It improves text generation efficiency through speculative decoding technology and performs well in mathematical reasoning and general Q&A.

Natural Language ProcessingTransformers

This is a Swahili-English translation model fine-tuned based on the opus-mt-mul-en model of Helsinki-NLP. The model is specifically optimized and trained for the children's hotline service scenario, trained with synthetic hotline dialogue data, equipped with a mechanism to prevent hallucination generation, and the translation quality is guaranteed by monitoring the BLEU score through an early stopping strategy.

Natural Language ProcessingTensorboardMultiple Languages

This is a large language model with 9B parameters based on the GLM architecture, specifically optimized for reasoning tasks. It uses the GGUF format for easy deployment and use in llama.cpp. The model has been processed to eliminate censorship and has an uncensored feature.

Natural Language ProcessingTransformersMultiple Languages

This is a quantized version based on the Huihui-Hunyuan-MT-Chimera-7B model, aiming to make large language models easier to deploy and use through quantization technology, enabling knowledge to be accessible to everyone.

Natural Language ProcessingGguf

This is a quantized version of the moonshotai/Kimi-K2-Instruct-0905 model, dedicated to making knowledge accessible to everyone. This project provides optimized model weights for easy deployment and use on various hardware.

Natural Language ProcessingGguf

The Ktiseos Nyx September Checkpoint is a text-to-image generation model based on LoRA fusion technology, licensed under the creativeml-openrail-m license. This model is built on the OnomaAIResearch/Illustrious-xl-early-release-v0 base model and trained using the EarthnDusk/XL_PDXL_Embeddings dataset, focusing on high-quality image generation tasks.

Computer VisionDiffusersEnglish

FastVLM is a lightweight vision-language model developed by Apple. It has 500 million parameters and supports text generation and vision-language understanding tasks. The model offers multiple quantization versions and can be easily run using the gguf-connector tool.

MultimodalGguf

This is the quantized version of ByteDance-Seed/cudaLLM-8B, which is committed to making large language models easier to deploy and use through model compression technology, realizing the concept of 'Let knowledge be used by everyone'.

Natural Language ProcessingGguf

TildeOpen LLM is an open-source foundational language model specifically designed to serve underrepresented Nordic and Eastern European languages. Funded by the European Commission and trained on the LUMI supercomputer, this model has 30 billion parameters and addresses the performance gap faced by users of 19 key languages in existing AI systems.

Natural Language ProcessingTransformersMultiple Languages

Marvis is an advanced conversational voice model designed for real-time streaming text-to-speech synthesis. It focuses on efficiency and ease of use, supporting high-quality real-time voice synthesis on consumer devices such as Apple chips, iPhones, iPads, and Macs.

Audio ProcessingTransformersEnglish

This is the Llama.cpp mixed layer quantization version of GLM-4.5-Air by the zai team. By adopting different quantization levels for each layer, a good balance is achieved between inference performance and model file size. Multiple quantization schemes are provided, suitable for running on machines with 64GB RAM.

Natural Language ProcessingGgufOther

This is a portrait generation model trained based on Qwen/Qwen-Image, specifically designed to generate anime images in the ancient Chinese style. The model is fine-tuned using LoRA technology and can generate digital artworks with traditional Eastern aesthetic features.

Computer VisionDiffusersEnglish

This is a computer vision model based on the YOLOv8s architecture, specifically designed for barcode and QR code detection. The model has been fine-tuned on a comprehensive dataset containing more than 5000 images, supporting accurate detection and classification of multiple barcode types (such as EAN13, Code128, etc.) and QR codes.

A quantized version of NVIDIA OpenReasoning - Nemotron - 32B, quantized through llama.cpp to reduce model storage and computational resource requirements for easy deployment.

Natural Language ProcessingGguf

This is a text generation model based on the EAGLE-3 speculative decoding algorithm, designed specifically for use with meta-llama/Llama-3.1-8B-Instruct. It improves text generation efficiency through speculative decoding technology and supports multiple languages.

Natural Language ProcessingSafetensorsMultiple Languages