marschr opened a new pull request, #17545: URL: https://github.com/apache/tvm/pull/17545
This PR adds the `powerPreferece` argument to `detetectGPUDevice()`, while keeping the previous functionality as default, without breaking older code. In hybrid GPU systems (mainly laptops) is sometimes needed to select the low-power integrated GPU in order to test model inference on https://github.com/mlc-ai/web-llm/ for other low-power devices or just to save on battery. Here's the related line of code on the `web-llm` repository that references this function: https://github.com/mlc-ai/web-llm/blob/082f04e4941ff4f6ef70731d244c69228948c7a1/src/service_worker.ts#L117 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
