"CoreWeave is an anchor provider of compute infrastructure for our transcoding network. The ability to Right-Size our workloads across CoreWeave’s diverse infrastructure set allows us to substantially reduce our compute costs and pass those savings along to our clients."

“Leveraging hardware capabilities introduced in the NVIDIA H100 GPU, such as thread block clusters, as well as new versions of CUTLASS, allows us to maintain the same Llama 2 70B cost per token generation that we achieve on NVIDIA A100 GPUs, while showing an almost 3x improvement in latency.”