Intel Software Optimizations Boost AI Inference in MLPerf V6.1
Intel, Wednesday, September 16th, 2026
Intel reports MLPerf Inference v6.1 gains driven mainly by software optimization rather than new silicon.
Intel published MLPerf Inference v6.1 results showing improved throughput and latency across its AI platforms.
The company attributes much of the gain to software stack work, including kernel optimization, quantization improvements, and better graph scheduling, on existing hardware. Results are broken down by benchmark workload and deployment scenario.
Intel frames the submission as evidence that software maturity is now a primary driver of inference economics.