Flux Copilot Can Now See

Now, not only can you chat with Copilot, but you can also show it what you're working on. What does that mean? Copilot is now the first multi-modal AI for hardware design, and with that, a whole new world of possibilities has opened up for Flux users.

Introducing: Image Input Capability!

Hardware design can be tricky because so much of the thinking is visual and requires context. As an engineer, you’re expected to rely on visual resources like block diagrams, charts, and drawings. But these resources are usually confusing to interpret and separate from your actual design, making them lose their value quickly.

With Copilot Vision, you can now provide Copilot with images as a more natural way to communicate ideas and integrate those resources into your design. Imagine snapping a photo of a block diagram and Copilot instantly understanding what you're trying to build.

Simply provide Copilot an image as a file upload and ask your questions. Suddenly, visual resources are relevant and even powerful for sanity checks. Now, you can ask questions, learn, and get design reviews in entirely new and more effective ways.

Unlocking New Workflows

Imagine you’re working on a project that’s based on an existing block diagram. You need to turn that block diagram into a working circuit, selecting the right components and connecting them in the right way. Normally this would mean reading through datasheets and application notes, but that gets confusing, time-consuming, and hard to keep track of.

Now that Copilot has vision, you have access to powerful new use cases that improve your hardware design workflow.

Part Recommendations: Trying to find the right parts to implement your design? You can now input a block diagram to Copilot and watch it recommend suitable parts for your design by intelligently parsing the diagram into functional sections.

‍Design Review: Once you make your design, you’ll want to make sure you didn’t miss anything. Copilot can help ensure design quality by comparing your schematic diagram against a block diagram. Copilot's analysis catches discrepancies, such as missing elements in your design that appear in the diagram, and offers you suggestions for improvement.

Chart Interpretation: Confused about how to interpret a chart on a datasheet? Just provide Copilot with an image of the chart and ask it in-depth questions. It will interpret your input and explain any aspect of the chart and how it relates to your hardware design needs.

‍Dimensional Analysis for Footprints: Have you ever looked at the part drawings and been overwhelmed by all the information and dimensions? Now you can provide an image to Copilot and ask it any question about part dimensions. Copilot efficiently interprets and correlates them to real component footprints, even making educated guesses about component types based on project context.

More Down the Road

This is a major step forward for Copilot, but we’re just scratching the surface of what’s possible. We believe that vision unlocks the doors to so much more down the road, but we’ll need your help! By trying out Copilot with vision, you can help us identify new use cases and steer the direction of Copilot’s future.

Try out Copilot today, and share your feedback with us through our Slack community. Tell us what works well, what could be better, and what other modes of interaction would you want to see us build into Copilot.