Brief-details: BART-large: A powerful transformer-based seq2seq model by Facebook, pre-trained for text generation and comprehension tasks with 149K+ downloads
Brief Details: SDXL-based LoRA model for generating Apple-style emoji artwork, featuring Pivotal Tuning and custom token embeddings. 179 likes, 2.8K+ downloads.
Brief-details: Bilingual Chinese-English dialogue model with enhanced capabilities including instruction-tuning, human feedback, and chain-of-thought improvements. Lightweight yet powerful.
Brief Details: A versatile text-to-image model trained on 10k high-quality public domain artworks, offering improved output quality across multiple styles with commercial restrictions.
Brief Details: Powerful 22B parameter code-focused model with multiple GGUF quantizations, optimized for programming tasks. Features various compression levels from 6.64GB to 23.64GB.
Brief Details: Advanced anime/3D art generation model with realistic lighting and textures. Optimized for high-quality character rendering with 5.3K+ downloads.
Brief Details: GLM-4-9B-Chat-1M is a powerful 9.48B parameter language model supporting 1M context length, 26 languages, and advanced features like function calling and web browsing.
Brief-details: IDEFICS-80B-instruct is an 80B parameter multimodal model fine-tuned for instruction following, capable of processing interleaved image-text inputs for diverse tasks like VQA and image captioning.
Brief-details: A 1.6B parameter instruction-tuned LLM by Stability AI, optimized via DPO, scoring 5.42 on MT-Bench. Notable for compact size and strong performance.
Brief-details: Open-Sora is an open-source video generation model that can create high-quality videos efficiently, with recent updates including 3D-VAE and score conditioning capabilities
Brief Details: Microsoft's Florence-2-base is a 0.23B parameter vision foundation model supporting multiple tasks like captioning, detection, and OCR with superior zero-shot capabilities.
BRIEF-DETAILS: FilmPortrait - A LoRA model fine-tuned on FLUX.1-dev that creates film-like images with Japanese cinema aesthetics, featuring grain textures and low saturation.
Brief Details: InternLM-XComposer2.5 - A 7B parameter vision-language model achieving GPT-4V level capabilities with support for 96K context length and multi-modal understanding.
BRIEF-DETAILS: A specialized Stable Diffusion model for pixel art generation, offering two distinct styles: pixelsprite and 16bitscene, with creative ML open rail license.
Brief-details: Advanced anime-style text-to-image model built on SDXL, featuring high-quality image generation with 170k+ training images and specialized LoRA adapters for style customization.
Brief-details: Optimized ONNX version of Phi-3-mini for accelerated inference, supporting 128K context length with int4/fp16 variants for CPU/GPU deployment
Brief-details: Yi-9B is a powerful 8.83B parameter language model excelling in code, math, and reasoning tasks, outperforming similar-sized models like Mistral-7B and SOLAR-10.7B.
Brief-details: A ControlNet model for Stable Diffusion that enables precise brightness control and grayscale image colorization, with over 5.8K downloads and CreativeML OpenRAIL-M license.
Brief Details: BELLE-7B-2M is a Chinese-English language model fine-tuned on Bloomz-7b1-mt with 2M instructions, optimized for text generation and understanding
Brief Details: FLUX.1-merged is a text-to-image diffusion model that combines FLUX.1-dev and FLUX.1-schnell models for efficient image generation in just 4 steps.
Brief-details: Advanced text-to-image model combining CLIP and Latent Diffusion, featuring multi-modal architecture with 3.27B+ parameters across components. Apache 2.0 licensed.
