Intel today announced their August 2025 Software Update to Project Battlematrix and the release of the LLM-Scaler 1.0 container for optimized AI inference support on Intel Arc B-Series graphics hardware.
Back in May Intel announced Project Battlematrix for supporting up to eight Intel Arc Pro GPUs for AI inference, SR-IOV support, improved vLLM performance, and a variety of other new features for their graphics stack on Linux. Their goal at the time was to have Q3 availability and full feature enablement by the end of the year.
Today they announced the LLM Scaler 1.0 container release for furthering this effort. LLM-Scaler 1.0 is described as:
“A new containerized solution built for Linux environment, optimized to deliver incredible inference performance with multi-GPU scaling and PCIe P2P data transfers, and designed to include enterprise-class reliability and manageability features such as ECC, SRIOV, telemetry and remote firmware updates.”
The 1.0 release incorporates new vLLM performance optimizations, various new vLLM features, better multi-modal model support, and more. The container also has oneCCL benchmaark support, XPU manager integration for various GPU telemetry features, and other enhancements.
More details on the Intel LLM-Scaler 1.0 container via GitHub.
Within the Intel.com announcement for their August 2025 software update they note up next will be a hardened version of LLM Scaler and other features by the end of Q3. The full feature release is still on track for the fourth quarter.
