A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 13:30 UTC
[CI/Build] workaround for CI build failure (#17070)
Signed-off-by: csy1204 <josang1204@gmail.com> Co-authored-by: Michael Goin <mgoin64@gmail.com>
Sangyeon Cho committed
Apr 24, 2025 at 08:14 UTC
b07d741661570ef199ba92528b11c0bd1294fb15
1 file changed
+6
docker/Dockerfile
+6
@@ -162,6 +162,9 @@ ENV UV_HTTP_TIMEOUT=500
162
COPY requirements/lint.txt requirements/lint.txt
163
COPY requirements/test.txt requirements/test.txt
164
COPY requirements/dev.txt requirements/dev.txt
165
+# Workaround for #17068
166
+RUN --mount=type=cache,target=/root/.cache/uv \
167
+ uv pip install --system mamba-ssm==2.2.4 --no-build-isolation
168
RUN --mount=type=cache,target=/root/.cache/uv \
169
uv pip install --system -r requirements/dev.txt
170
#################### DEV IMAGE ####################
@@ -265,6 +268,9 @@ ADD . /vllm-workspace/
268
ENV UV_HTTP_TIMEOUT=500
269
270
# install development dependencies (for testing)
271
+# Workaround for #17068
272
+RUN --mount=type=cache,target=/root/.cache/uv \
273
+ uv pip install --system mamba-ssm==2.2.4 --no-build-isolation
274
RUN --mount=type=cache,target=/root/.cache/uv \
275
uv pip install --system -r requirements/dev.txt
276