This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Seriously Intelligent Software
Mapping trained AI models to silicon without extensive rework has historically been challenging. Deep Vision’s software development kit turns this dynamic around to make the entire process transparent and friction-free for the customer. The result is a low code, automated, virtually seamless migration process from the training AI models to production deployment. The tool suite reduces expensive development time by dramatically increasing productivity and reducing overall time to market.
Seamless Porting
Trained Neural Network Models
Quanitizer
During conversion of AI models, automatically selects layers to represent in 8-bit or 16-bit integer notation, ensuring highest model accuracy.
Profiler
Delivers layer by layer performance statistics of AI models to discover and unlock bottlenecks.
Optimizer
Optimize runtimes to fit application needs such as dialing the overall power consumption to fit system power requirements
Simulator
Determine power, performance and accuracy of models by running on them on Deep Vision’s bit accurate simulator before deploying them on Silicon
Industry Standard Frameworks
Supports industry-standard frameworks and networks, offering clear advantages in migrating new applications into production.
Want to Know More?
Deep Vision provides world-class tools to develop and deploy powerful Edge AI applications successfully.
Download SDK Brief