Sponsors

d-Matrix is building a next generation compute platform for Generative AI workloads focused on inference at-scale. The d-Matrix compute architecture uses ground-breaking digital in-memory-compute cores and chiplet interconnect for faster throughput, lower latency, and better economics when running Generative AI inference compared to alternative solutions. The d-Matrix hardware is supported by an enterprise-grade software suite using open ecosystem tools for easy deployment of Generative AI models. Users can easily optimize, compile, and distribute workloads across the d-Matrix hardware with minimal model changes and without re-training. d-Matrix is transforming AI from unsustainable to attainable.