Brief-details: A powerful 7.11B parameter text embedding model from Salesforce, built on Mistral-7B, optimized for retrieval tasks with MTEB benchmark performance
BRIEF DETAILS: 8B parameter LLaMA-3 model fine-tuned for Russian language, achieving GPT-3.5-turbo level performance. Optimized for GGUF format with 73K+ downloads.
BRIEF-DETAILS: A Chinese language sentence transformer with variable dimension embeddings (128-1792d), optimized for retrieval and semantic tasks with 326M parameters. Achieves strong CMTEB benchmark scores.
Brief Details: LLaVA-NeXT 7B - Advanced multimodal vision-language model with improved OCR and reasoning capabilities, 7.06B parameters, FP16 precision
Brief Details: BigBird-RoBERTa base model - Transformer-based architecture supporting 4096-length sequences using block sparse attention. Apache 2.0 licensed.
Brief-details: SD-Turbo is a fast text-to-image model by StabilityAI that generates high-quality images in a single step, distilled from Stable Diffusion 2.1 using Adversarial Diffusion Distillation.
Brief-details: A PyTorch-based neural vocoder for high-quality audio synthesis, converting acoustic features to waveforms using GAN and Fourier transforms at 24kHz
Brief Details: OWLv2 is a 438M parameter zero-shot object detection model using CLIP backbone with ViT-L/14 architecture, enabling text-conditioned object detection.
Brief-details: Multilingual late interaction retriever supporting 94 languages with 559M params. Features Matryoshka embeddings and superior retrieval performance compared to v1.
Brief-details: AnimateLCM is a computation-efficient text-to-video generation model capable of creating high-quality animated content in just 4 steps, offering fast inference with personalized styling.
Brief Details: A lightning-fast text-to-video generation model by ByteDance that runs 10x faster than original AnimateDiff with 1-8 step options.
Brief Details: Japanese medical NER model (110M params) for identifying medical entities in clinical text. Supports disease, medication & temporal annotations.
Brief Details: Powerful multilingual speech model with 1B parameters, supporting 128 languages. Pre-trained on 436K hours of speech data, ideal for ASR tasks.
Brief-details: Arabic Named Entity Recognition model built on CAMeLBERT-Mix, fine-tuned on ANERcorp dataset for accurate entity detection in Arabic text.
Brief-details: A fine-tuned DistilRoBERTa model for NSFW text classification, trained on 14,317 Reddit posts to detect inappropriate content with binary classification (NSFW/SFW).
BRIEF-DETAILS: ESM-2 protein language model with 150M parameters. Features 30 layers, MIT license, optimized for masked language modeling of protein sequences.
Brief Details: Photorealistic text-to-image model optimized for generating high-quality images, especially of people. Features 840KVAE integration and CreativeML license.
Brief-details: TimeSformer base model fine-tuned on Kinetics-400 dataset for video classification, implementing space-time attention mechanisms with transformer architecture.
Brief Details: NFNet-L0 is a lightweight, normalization-free neural network with 35.1M parameters, optimized for ImageNet classification using scaled weight standardization.
Brief-details: Efficient cross-encoder model for MS Marco passage ranking, achieving NDCG@10 of 67.43 on TREC DL 19, processing 9000 docs/sec on V100 GPU
BRIEF-DETAILS: A 13B parameter LLaMA2-based creative writing model optimized for storytelling, chatbots, and adventures, featuring merged capabilities from multiple specialized LORAs
