Tartan AI is a hardware/firmware company that works on optimizations for machine learning workloads. It delivers building blocks for hardware acceleration for deep learning, offering processing elements and transparent memory compression technologies.
The company’s portfolio covers processing elements and transparent memory compression technologies that reduce the cost of data storage and transfers, in addition to computations when performing multiply-accumulate operations, enabling semiconductor companies to boost performance and energy efficiency by reducing the work, storage and communication needed when executing machine learning models in silicon.
Tartan AI’s IP portfolio covers technologies with:
- Profiling tools for enhancing sparsity
- Profiling/training methods for reducing data width
- Processing elements that support variable width data types up to the full spectrum at bit resolution, 1b to 16b in 1b or 2b steps
- Bit-Skipping processing elements that skip all computations with zero bits
- Specialized solutions for computational imaging taking advantage of spatial value locality
Website: https://www.tartanai.com/