llama: only use one iGPU device by default#23897

0cc4m merged 1 commit into

masterggml-org/llama.cpp:masterfrom

0cc4m/igpu-deduplicationggml-org/llama.cpp:0cc4m/igpu-deduplicationCopy head branch name to clipboard