Brief-details: A Q8-quantized 27B parameter Gemma model optimized for llama.cpp, converted from VIDraft/Gemma-3-R1984-27B to GGUF format for efficient local deployment
Brief-details: Qwen2.5-VL-32B-Instruct GGUF conversion - 32B parameter multimodal LLM optimized for Q4 quantization, supporting visual and language tasks via llama.cpp
Brief-details: AccVideo is a high-performance video diffusion model that achieves 8.5x faster inference than HunyuanVideo, utilizing synthetic datasets for efficient distillation.
Brief-details: A 32B parameter quantized GGUF model converted from VIDraft/QwQ-R1984-32B, optimized for llama.cpp deployment with Q4_K_M quantization format
Brief-details: Quantized 12B parameter Gemma model converted to GGUF format for efficient local inference using llama.cpp, optimized for Q6_K precision
Brief-details: A GGUF-formatted 12B parameter Gemma model optimized for llama.cpp, featuring 8-bit quantization and enhanced inference capabilities for efficient deployment
Brief-details: Gemma3-R1984-12B is a 12B parameter multimodal AI platform built on Google's Gemma architecture, featuring web search integration, 8K token context, and secure local deployment capabilities.
BRIEF DETAILS: A LoRA model for generating stylized images of Morgenstern with different personas, featuring high-resolution 26K output and customizable hair colors.
Brief-details: A LoRA model for generating varied Boris Yeltsin images with different hair colors and roles, including Japanese president and police officer positions.
Brief-details: A specialized LoRA model for generating black string sandal images, built on FLUX.1-dev base model. Requires "SS" trigger word for optimal results.
BRIEF-DETAILS: A LoRA model specialized in generating images of Asian females wearing specific necklace designs (NL1), optimized for outdoor photography scenarios with natural lighting
Brief Details: Eastern aesthetic-focused AI model for portraits and scenes, specializing in diverse age groups, cultural elements, and environmental representations. Weight: 0.9-1 recommended.
BRIEF-DETAILS: State-of-the-art 7B parameter code embedding model supporting 6 programming languages, outperforming competitors on CodeSearchNet with advanced retrieval capabilities
Brief Details: BizGen is an advanced AI model for generating business infographics and slides from article-level text, featuring ultra-dense layouts and multilingual text rendering capabilities.
Brief-details: Mirage-Ghibli is a specialized image generation model that creates Studio Ghibli-style artwork, accessible through miyazakify.com using the GHBLI trigger word.
Brief-details: Advanced image-to-3D generation model using rectified flow transformers with 1.5B parameters, capable of producing high-fidelity 3D meshes from single images.
Brief-details: MegaTTS3 is ByteDance's advanced text-to-speech model available on HuggingFace, designed for high-quality speech synthesis applications.
BRIEF-DETAILS: A 1.3B parameter video control model supporting multiple conditions (Canny, Depth, Pose, MLSD) with multi-resolution capabilities and trajectory control at 16fps.
Brief Details: A Ghibli-style LoRA model for FLUX.1-dev that generates Studio Ghibli-inspired artwork, specializing in fantasy-mechanical fusion scenes with Japanese aesthetics
Brief Details: A powerful 27B parameter multimodal AI model built on Google's Gemma, featuring web search integration, 8K token context, and secure local deployment on NVIDIA A100 GPUs.
BRIEF-DETAILS: DeepSeek-V3-0324 is an advanced language model with improved reasoning, web development, and Chinese capabilities, offering multiple quantization options from 1.78-4.5 bits for efficient deployment.
