AI chip designer Blaize has launched an end-to-end ‘no code’ tool for developing machine learning computer visions for edge AI.
AI Studio integrates a number of tools and workflows with a simplified interface. An API will allow other tools to be added to the workflow and the company is accelerating plans for security elements.
The tool aims to simplify the development and deployment of AI frameworks, optimising them for the Blaize Graph Streaming Processor (GSP). This would work separately to the Picasso software development kit (SDK) that is aimed at software engineers working on computer vision algorithms.
“We saw our early hardware customers facing barriers to operation at the edge due to lack of software tools,” said Dinakar Munagala, Co-founder and CEO at Blaize. “Today in edge computing there’s a huge dearth of tools to build AI applications. While AI is migrating to the edge and outpacing the data centre, the deployment is lagging as it takes too long to build the apps and deploy the apps to hardware.”
The drag-and-drop workflow is deliberately using open standards, says Dmitry Zakharchenko, VP Research & Development at Blaize. Developers can deploy models with one click using ONNX, OpenVX, containers, Python, or GStreamer. The tool also gives access to public and private libraries and repositories to add in software components and even complete computer vison applications.
“AI development today is very complicated. Even containers with preloaded libraries for GPU are not suitable for production,” said Zakharchenko. “The workflow is transparent, at any stage you can see what Studio has done, and it uses AI models in ONNX and openVX, we do not like black boxes.”
“AI studio is tightly integrated with Blaize hardware but its based on open standards so we can deploy to any hardware that supports open standards such as ONNX,” he said. “We have developed a robust set of APIs to