The Phi-4-mini-instruct model undergoes float8 dynamic activation and weight quantization via torchao, achieving 36% VRAM reduction and 15-20% speed improvement on H100 with minimal accuracy impact.

Natural Language ProcessingTransformersOther

LoRA trained on the Wan2.1 14B I2V 480p model, capable of transforming images into videos with Fus-Ro-Dah spell impact effects

Computer VisionDiffusersEnglish

This model is used to analyze the impact and transformation trends of artificial intelligence technology on various social sectors

Natural Language ProcessingSafetensorsJapanese

A version fine-tuned for 1 day using the Docmatix dataset (5% data volume) based on Microsoft's Florence-2 model, suitable for image-text understanding tasks

MultimodalTransformers

This model uses vibrant flat colors and strong contrasts to create highly impactful illustrations, blending modernity with retro charm.

Computer VisionDiffusers

A line art portrait illustration generation model based on Stable Diffusion XL, excelling in creating impactful monochrome manga-style images

Computer VisionDiffusers

WhiteRabbitNeo is an AI model series focused on cybersecurity, with the 33B version demonstrating its capabilities in offensive and defensive security and assessing social impact.

Natural Language ProcessingTransformers

This is a laser-optimized version of senseable/WestLake-7B-v2. By applying the laserRMT method to perform laser intervention on specific layers, it aims to reduce the impact of noise and improve the model's performance. It is sponsored by VAGO Solutions and HyperSpace.Ai.

Natural Language ProcessingTransformers

Sheared-LLaMA-1.3B-Pruned is a 1.3B parameter model pruned from Llama-2-7b, without continued pretraining, primarily used for studying pruning techniques and their impacts.

Natural Language ProcessingTransformers

A version fine-tuned using the IFprompMKR dataset with qlora on the microsoft/phi-1_5 model, primarily for text generation tasks.

Natural Language ProcessingTransformers

A GGUF format model based on the 7B version of llama2, designed for prompt generation and compatible with CPU environments

Natural Language ProcessingGguf

2D-Mix is a text-to-image generation model based on stable diffusion technology, specifically optimized for anime style and Genshin Impact characters.

Computer VisionDiffusersEnglish

OPT is an open-source large language model series launched by Meta AI, benchmarked against the GPT-3 architecture, aimed at promoting reproducibility and societal impact discussions in large model research.

Natural Language ProcessingJaxEnglish

A RoBERTa model employing pre-layer normalization technology, studying the impact of masking ratio in masked language modeling

Natural Language ProcessingTransformers

A masked language model based on the RoBERTa architecture, employing pre-layer normalization technology to explore the impact of masking ratios on model performance

Natural Language ProcessingTransformers