From 2bd0845aa680e77e3ae2a633cc7a6736f32afadc Mon Sep 17 00:00:00 2001
From: Gilbert Gong <gilbert.gong@gmail.com>
Date: Thu, 31 Oct 2024 20:46:00 -0700
Subject: [PATCH] llama.cpp-base: compile just the binaries we need due to
 github runner disk space

---
 llama.cpp-base/Dockerfile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/llama.cpp-base/Dockerfile b/llama.cpp-base/Dockerfile
index 87edab2..b0ea175 100644
--- a/llama.cpp-base/Dockerfile
+++ b/llama.cpp-base/Dockerfile
@@ -34,7 +34,7 @@ ENV LLAMA_CURL=1
 ENV LLAMA_CUDA_FA_ALL_QUANTS=1
 ENV GGML_CUDA_FA_ALL_QUANTS=1
 
-RUN make -j 12
+RUN make -j 12 llama-batched-bench llama-server
 
 FROM ${CUDA_BASE_CONTAINER} as runtime