Update build docs for --image,gpu-base (triton-inference-server#4209)

dhruv2180 · Apr 12, 2022 · f7f3262 · f7f3262
1 parent 2a94cfa
commit f7f3262
Showing 1 changed file with 5 additions and 0 deletions.
diff --git a/docs/build.md b/docs/build.md
@@ -131,6 +131,11 @@ To include the TensorFlow2 backend in your CPU-only build, you must
 provide this additional flag to build.py:
 `--extra-backend-cmake-arg=tensorflow2:TRITON_TENSORFLOW_INSTALL_EXTRA_DEPS=ON`.
 
+When building without GPU support, you must use the `--image=gpu-base,nvcr.io/nvidia/tritonserver:<xx.yy>-py3-min`
+flag. This is needed since the CPU-only builds of the TensorFlow and
+PyTorch backends require some CUDA stubs and runtime dependencies that are
+not present in the CPU-only base container.
+
 ### <a name="ubuntu-without-docker"></a>Building without Docker
 
 To build Triton without using Docker you must install the build