Benchmark results and performance data for the Intel Arc Pro B70 GPU (Xe2/Battlemage) - LLM inference, video generation, dual-GPU scaling.
-
Updated
Apr 22, 2026
Benchmark results and performance data for the Intel Arc Pro B70 GPU (Xe2/Battlemage) - LLM inference, video generation, dual-GPU scaling.
Field-tested guide: multi-GPU vLLM tensor-parallel (TP=2/TP=4) on Intel Arc Pro B70 (Battlemage BMG-G31, Xe2) on Linux. Driver setup (xe force_probe=e223), bare-metal vLLM + oneAPI 2025.3, the compute-runtime multi-root USM + triton-xpu init_devices fixes, FP8/int4-AutoRound quant, root-cause error reports. AI-agent readable (AGENTS.md).
Brunch fork adding Intel Lunar Lake (Xe2) support: 7.1 kernel, Mesa 25.3.6 userspace, ARCVM graphics fixes. Upstream: sebanc/brunch-unstable
Registration and management system for university courses
Xe2 dual-B70 kernel and 2x2 parallelism lab (TP=2, PP=2)
Pioneer documentation: OpenFOAM CFD GPU offloading on Intel Arc Pro B70 Pro (Battlemage Xe2). Hardware excellent (FP64 96%, kernel latency CUDA-par), software stack not ready (GPU 66% idle, no working strong preconditioner in Ginkgo 1.10 SYCL). 14 bugs documented + direct VRAM/profiling data.
Native Intel Xe2 LLM inference: SYCL/DPAS kernels, paged KV, prefix caching, continuous batching, and XMX packed prefill
To associate your repository with the xe2 topic, visit your repo's landing page and select "manage topics."