官方定位为 tiny 的视觉语言模型,主打小而够用:在普通设备上完成看图问答、目标定位、OCR 与界面元素识别,适合边缘端部署和批量图像处理。仓库提供本地推理、微调与评测脚本。 ⭐ GitHub Stars: 10k+ | 🔗 仓库: https://github.com/m87-labs/moondream
A vision language model the project positions as tiny, small enough to run on ordinary hardware while still handling image Q&A, pointing, OCR and UI element detection, which makes it practical for edge deployment and bulk image processing. The repo ships local inference, fine-tuning and evaluation scripts. ⭐ GitHub Stars: 10k+ | 🔗 Repo: https://github.com/m87-labs/moondream
Moondream belongs to the Open Source LLMs category on hedirbase. Tagged with: vision-language-model,multimodal,edge-ai,ocr,open-source.
A vision language model the project positions as tiny, small enough to run on ordinary hardware while still handling image Q&A, pointing, OCR and UI element detection, which makes it practical for edg