# Efficient inference with 7 billion parameters

Quantized version of Qwen2.5-VL-7B-Abliterated-Caption-it, supporting multilingual image description tasks.

The Nunchaku quantized version of the Colossus Project Flux, designed to generate high-quality images based on text prompts. This model minimizes performance loss while optimizing inference efficiency.

This is a static quantized version based on the Qwen2.5-VL-7B model, focusing on image captioning generation tasks and supporting multiple languages.

olmOCR-7B-0725-FP8 is a document OCR model based on the Qwen2.5-VL-7B-Instruct model. It is fine-tuned using the olmOCR-mix-0225 dataset and then quantized to the FP8 version.

Lucy-128k is a model developed based on Qwen3-1.7B, focusing on proxy-based web search and lightweight browsing, and can run efficiently on mobile devices.