Intel Achieves First, Only Full NPU Support in MLPerf Client v0.6 Benchmark

intel achieves first, only full npu support in mlperf client v0.6 benchmark. may 5, 2025 published client computing results show intel core ultra series 2 processors offer unprecedented ai compute performance spanning the cpu, gpu and npu. in this article: what’s new: intel today announced that it is the only company to achieve full neural processing unit (npu) support in the newly released mlperf client v0.6 benchmark. the result marks the industry’s first standardized evaluation of large language model (llm) performance on client npus. intel’s measurements of mlperf client v0.6 show intel® core™ ultra series 2 processors can produce output on both the graphics processing unit (gpu) and the npu much faster than a typical human can read. “we are proud to lead the industry in enabling full npu acceleration and industry-leading gpu performance for ai workloads on client pc platforms. this success reflects intel’s deep hardware-software co-optimization and commitment to democratizing ai for pcs everywhere." –daniel rogers, intel vice president and general manager of pc product marketing. why it matters: with its intel core ultra series 2 processors, intel is at the forefront of the ai pc evolution, offering unprecedented ai compute performance spanning the central processing unit (cpu), gpu and npu. mlperf client v0.6 measures four content generation and summarization use cases based on the llama 2 7b model. intel demonstrated leading performance across npu and built-in intel® arc™ gpu. intel achieved the fastest npu response time, generating the first word in just 1.09 seconds (first token latency), meaning it begins answering almost immediately after receiving a prompt. it also delivered the highest npu throughput at 18.55 tokens per second, referring to how quickly the system can generate each additional piece of text, enabling seamless real-time ai interaction. additionally, compared to competition, intel showed gpu leadership in time to first token, starting faster than the competition and reinforcing its npu and gpu end-to-end ai acceleration advantage. about npu benchmarking on mlperf: developed collaboratively by mlcommons consortium members — including intel, amd, microsoft, nvidia and qualcomm — mlperf client v0.6 extends beyond previous gpu-centric tests to now include dedicated npu benchmarking. driven by close collaboration between intel's npu hardware and openvino™ software teams, intel core ultra processors remain the only npu to achieve complete npu compliance in the final benchmark. more: press kit: intel core ultra processors (series 2). editor's note: gpu results and cpu model number were updated after original publication on may 5, 2025. the small print:. testing configuration. amd intel oem platform asus zenbook s 16 asus zenbook s 14 oem model number um5606wa ux5406sa cpu model amd ryzen ai hx 370 intel® core™ ultra 9 288v processor bios date march 21, 2025 february 26, 2025 bios version um5606wa.317 ux5406sa.306 total memory 32gb lpddr5, 7500 mhz 32gb lpddr5, 8533 mhz graphics brand amd radeon 890m intel arc 140v storage memory 1tb 1tb os windows 11 pro x64 windows 11 pro x64 power source ac ac power plan balanced balanced power mode best performance best performance oem power setting myasus: fullspeed myasus: fullspeed * all data measured as of april 28, 2025. see press kit for workload and configuration details. notices & disclaimers. performance varies by use, configuration and other factors. learn more at intel.com/performance index. performance results are based on testing as of dates shown in configurations and may not reflect all publicly available updates. see backup for configuration details. no product or component can be absolutely secure. your costs and results may vary.