Try the Model

Use the widget below to experiment with BLIPv2. You can detect COCO classes such as people, vehicles, animals, household items.

Overview

BLIPv2 is a multimodal model developed by Salesforce Research. You can use BLIPv2 for visual question answering and zero-shot image classification. BLIPv2 supersedes BLIP, offering superior performance.

BLIPv2 License

BLIPv2

is licensed under a

license.

Performance

Deploy a BLIPv2 API

You can use Roboflow Inference to deploy a

BLIPv2

API on your hardware. You can deploy the model on CPU (i.e. Raspberry Pi, AI PCs) and GPU devices (i.e. NVIDIA Jetson, NVIDIA T4).

Below are instructions on how to deploy your own model API.

Label Data Automatically with BLIPv2

You can automatically label a dataset using BLIPv2 with help from Autodistill, an open source package for training computer vision models. You can label a folder of images automatically with only a few lines of code. Below, see our tutorials that demonstrate how to use BLIPv2 to train a computer vision model.

No items found.