Benchmarks: OpenCL Kernel Latency ~76x Lower For Intel Lunar Lake With Updated Compute Runtime
There were some nice improvements in other OpenCL workloads too when going from last month's Compute Runtime release to the newly-released Compute Runtime 25.18.33578.6 update.
Across more than 40 benchmarks, the geo mean improved by 14% with the new Intel Compute Runtime on Xe2 Lunar Lake integrated graphics. Quite a nice improvement overall and especially for any latency-sensitive tasks with the Ultra Low Latency Scheduling (ULLS) work for Lunar Lake. Very nice software improvements by Intel that can now be enjoyed with Lunar Lake on Linux.
If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.
