llama-cpp on intel-arc-pro-b70: Gemma 4 31B gguf-q3-k-m pp512 tok/s (bentech, single B70 SYCL)

THIRD-PARTY MEASURED (Ben Linville/bentech, Apr 16 2026, llama-benchy 0.3.5, 25-run matrix). Arc Pro B70 32GB (BMG-G31) + B60 24GB (BMG-G21); Ryzen 7 7800X3D, 64GB DDR5, B850 PCIe4 x4 M.2 adapters; llama.cpp b1-ff5ef82 Vulkan+SYCL, Intel llm-scaler-vllm:0.14.0-b8.1 Docker (vLLM FP8 pipeline-parallel). pp512/tg128, depth0, 3 runs/test mean, no prefix caching, q8_0 KV, FA on, 8k ctx. Per-request tg, NOT aggregate. Gemma 4 31B gguf-q3-k-m, single B70 SYCL.

pp512 174.7 tok/s prefill on single B70 SYCL.