From 0ad0e738cd7ed1266b3c210ad54dcd2b70142563 Mon Sep 17 00:00:00 2001 From: Richard Lyons Date: Sun, 18 Aug 2024 01:43:26 +0200 Subject: [PATCH] Override numParallel only if unset. --- server/sched.go | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/server/sched.go b/server/sched.go index 9947fd325..4d9c02969 100644 --- a/server/sched.go +++ b/server/sched.go @@ -734,7 +734,9 @@ func pickBestFullFitByLibrary(req *LlmRequest, ggml *llm.GGML, gpus gpu.GpuInfoL // If multiple Libraries are detected, pick the Library which loads the most layers for the model func pickBestPartialFitByLibrary(req *LlmRequest, ggml *llm.GGML, gpus gpu.GpuInfoList, numParallel *int) gpu.GpuInfoList { - *numParallel = 1 + if *numParallel <= 0 { + *numParallel = 1 + } byLibrary := gpus.ByLibrary() if len(byLibrary) <= 1 { return gpus