# Low VRAM requirement

Qwen3-8B-FP8-dynamic is an optimized version of the Qwen3-8B model through FP8 quantization, significantly reducing GPU memory requirements and disk space usage while maintaining the original model's performance.

HiDream-I1 is an open-source image generation foundation model with 17 billion parameters. The 4-bit quantized version can run on 16GB VRAM, enabling fast and high-quality image generation.

Wan2.1 is a comprehensive open-source video foundation model with top-tier performance, supporting consumer-grade GPUs, multi-task capabilities, visual text generation, and efficient video VAE.

Wan2.1 is an open and advanced large-scale video generation model that supports various tasks including text-to-video and image-to-video, compatible with consumer-grade GPUs.

This is the 4-bit quantized version of the mistralai/Mistral-Small-24B-Instruct-2501 model, quantized by ConfidentialMind.com, achieving a smaller and faster model with minimal performance loss.

INT4 quantized version of FLUX.1-schnell, enabling efficient text-to-image generation with SVDQuant technology

FP8 quantized version of Llama-3.2-1B-Instruct, suitable for multilingual business and research applications, with performance close to the original model.

FP8 quantized version of Meta-Llama-3.1-405B-Instruct, suitable for multilingual commercial and research purposes, specially optimized for assistant robot scenarios.

FP8 quantized version of Meta-Llama-3.1-8B-Instruct, suitable for multilingual business and research applications, specially optimized for assistant-like chat scenarios.

An efficient text-to-image generation model fine-tuned based on Stable Diffusion XL, supporting rapid generation of artistic images

SoteMix V2.1 is a high-resolution text-to-image model based on Stable Diffusion, specializing in artistic and anime-style image generation.

A text-to-image generation model fine-tuned from SSD-1B using LCM-LoRA technology, supporting rapid generation of high-quality images

A lightweight text-to-image generation model optimized through distillation based on Realistic_Vision_V4.0, achieving 80% faster speed than base SD1.5

LLaVA is a multimodal model combining vision and language, based on the LLaMA architecture, supporting image understanding and dialogue generation.

This is the 8-bit quantized version of EleutherAI's GPT-J 6B parameter model, optimized for running and fine-tuning on limited GPU resources (e.g., Colab or 1080Ti).