Brief-details: Large-scale speech recognition model with 1.54B parameters, supporting 99 languages. Achieves 3.0 WER on LibriSpeech. Handles both transcription and translation.
Brief Details: RWKV-4 Raven is a powerful RNN-based language model series (1.5B-14B params) with strong multilingual capabilities, optimized for text generation and chat applications.
Brief Details: High-quality video generation model capable of 1024x576 resolution, specializing in watermark-free video upscaling from zeroscope_v2_576w outputs.
Brief-details: A specialized comic art style diffusion model supporting 6 distinct artistic styles, perfect for comic creation with 506 likes and 1.4K+ downloads. Allows style mixing for unique results.
Brief-details: Advanced YOLOv8-based detection model for faces, hands, people and clothing, with multiple pre-trained variants offering high accuracy segmentation and detection capabilities
Brief-details: A powerful multilingual translation model supporting 96 languages for text and 101 for speech, enabling seamless speech-to-speech, speech-to-text, and text-to-text translation.
BRIEF DETAILS: 176B parameter multilingual language model fine-tuned on xP3 dataset, capable of following instructions in 46+ languages with strong zero-shot performance.
Brief Details: MPT-7B-Chat: 6.7B parameter chatbot model fine-tuned on multiple dialogue datasets. Features FlashAttention and ALiBi. Non-commercial license.
Brief-details: Advanced text-to-image model built on Stable Diffusion, featuring BLIP-2 integration and enhanced composition freedom. 517 likes, focuses on natural language processing.
Brief-details: Extended context LLaMA-2 variant with 7B parameters, supporting 32K token context length. Optimized for long-form tasks like document QA and summarization.
Brief-details: Collection of anime character LoRA models for stable diffusion, featuring 20+ characters from Genshin Impact, Blue Archive, and other games. High-quality character generation.
Brief-details: A specialized text-to-image diffusion model fine-tuned for generating cyberpunk anime characters, based on Waifu Diffusion V1.3 with Stable Diffusion V1.5 VAE.
Brief Details: Intel's 7B parameter LLM fine-tuned from Mistral-7B-v0.1, optimized for enhanced performance with 8192 token context window and DPO alignment
Brief-details: Llama3-ChatQA-1.5-8B is an 8B parameter model optimized for conversational QA and RAG, built on Llama-3 with enhanced capabilities for tabular and arithmetic calculations.
Brief-details: Instruction-finetuned text embedding model that generates task-specific embeddings through natural language instructions. SOTA on 70+ tasks.
Brief-details: A specialized ControlNet model trained on LAION Face Dataset for precise control of facial expressions and gaze direction in image generation, compatible with SD2.1 and SD1.5
Brief-details: A highly aesthetic text-to-image diffusion model generating 1024x1024 images, outperforming SDXL by 2.5x in user preferences, with FID score of 7.07.
Brief-details: A 30B parameter uncensored LLaMA-based model quantized to 4-bit GPTQ, offering high performance without alignment restrictions. Optimized for efficient GPU inference.
Brief-details: A specialized Stable Diffusion model fine-tuned for high-quality 3D artwork generation, featuring Cinema4D-inspired redshift rendering aesthetics and capabilities.
Brief-details: Anime-focused text-to-image diffusion model based on Stable Diffusion 1.4, fine-tuned on 680k high-quality anime images with multiple weight variants available.
Brief Details: Advanced AI model using 4-bit quantization with improved V2 version featuring better precision and faster inference through optimized compression techniques
