Florence-2-base is a multimodal foundation model hosted on Hugging Face that processes image and text inputs to generate text outputs. Microsoft released the model under the MIT license with open weights for use in research and application development. It belongs to the class of vision-language models implemented in the Transformers library.
The model card lists it as supporting the image-text-to-text task. Code examples demonstrate loading via the Transformers pipeline with the task specifier "image-text-to-text" or directly through AutoProcessor and AutoModelForMultimodalLM classes, both requiring trust_remote_code=True. An additional integration path with vLLM is shown through a pip installation command for serving the model.
It is delivered as a downloadable model repository containing PyTorch and Safetensors files along with custom code. The associated arXiv paper is referenced as 2311.06242. Developers access it through the Hugging Face ecosystem, including notebooks on Google Colab and Kaggle, and local applications.
The model carries an MIT license. No pricing information appears for the model itself.
Florence 2 Base sits in PulseGate's Foundation models & chat category. It enables developers to build and deploy advanced multimodal AI applications using open-source image and text models. Florence 2 Base is an open-source project aimed at AI researchers and developers. The project is open source (Open Source). The product ships for the command line, and it can be self-hosted.
It is developed by Microsoft, and the product first shipped in 2023. Among its 5 catalogued features are image-text modeling, transformer architecture, and pyTorch support.
Latest indexed changes and source events
microsoft/Florence-2-base verified by the PulseGate indexer
Other apps tracked under the same category.
