remove server repo from test

docs/user_guide/architecture.md

@@ -311,8 +311,9 @@ description of the model contains variable-sized dimensions, Triton will use *1*
 for every variable-sized dimension for the starting request. For other
 non-starting requests in the sequence, the input state is the output state of
 the previous request in the sequence. For an example ONNX model that uses
-implicit state you can refer to
-[this ONNX model](../../qa/common/gen_qa_implicit_models.py#L101).
+implicit state you can refer to this onnx model generated from the 
+`create_onnx_modelfile_wo_initial_state()`
+[from this generation script](../../qa/common/gen_qa_implicit_models.py).
 This is a simple accumulator model that stores the partial sum of the requests
 in a sequence in Triton using implicit state. For state initialization, if the
 request is starting, the model sets the "OUTPUT\_STATE" to be equal to the

qa/L0_doc_links/test.sh

@@ -21,16 +21,15 @@ python3 -m pip install mkdocs-htmlproofer-plugin
 
 # Get the necessary repos
 mkdir repos && cd repos
-TRITON_SERVER_BRANCH_NAME=${TRITON_SERVER_BRANCH_NAME:="main"}
 TRITON_BACKEND_REPO_TAG=${TRITON_BACKEND_REPO_TAG:="main"}
-echo ${TRITON_SERVER_BRANCH_NAME}
-git clone --single-branch --depth=1 -b ${TRITON_SERVER_BRANCH_NAME} https://github.com/triton-inference-server/server.git
 echo ${TRITON_BACKEND_REPO_TAG}
 git clone --single-branch --depth=1 -b ${TRITON_BACKEND_REPO_TAG} https://github.com/triton-inference-server/backend.git
 cd ..
 
 exec mkdocs serve -f $CONFIG > $LOG &
 PID=$!
+# Time for the compilation to finish. This needs to be increased if other repos
+# are added to the test
 sleep 20
 
 until [[ (-z `pgrep mkdocs`) ]]; do