huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-06-21 16:40:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

backend(trtllm): forget to bump dockerfile

Browse Source
This commit is contained in:
Morgan Funtowicz 2025-02-05 10:27:47 +01:00
parent 6168ffc23f
commit 4c44de4ee7
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
Dockerfile_trtllm
View File

@ -5,8 +5,10 @@ ARG sccache_gha_enabled=off
ARG actions_cache_url="" ARG actions_cache_url=""
ARG actions_runtime_token="" ARG actions_runtime_token=""
ENV CUDA_BASE=12.8.0
# CUDA dependent dependencies resolver stage # CUDA dependent dependencies resolver stage
FROM nvidia/cuda:12.6.3-cudnn-devel-ubuntu24.04 AS cuda-builder FROM nvidia/cuda:${CUDA_BASE}-cudnn-devel-ubuntu24.04 AS cuda-builder
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \ RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
build-essential \ build-essential \
@ -105,7 +107,7 @@ RUN export CMAKE_C_COMPILER_LAUNCHER=sccache && \
cargo build --profile ${build_type} --package text-generation-backends-trtllm --bin text-generation-backends-trtllm && \ cargo build --profile ${build_type} --package text-generation-backends-trtllm --bin text-generation-backends-trtllm && \
sccache --show-stats sccache --show-stats
FROM nvidia/cuda:12.8.0-cudnn-runtime-ubuntu24.04 AS runtime FROM nvidia/cuda:${CUDA_BASE}-cudnn-runtime-ubuntu24.04 AS runtime
RUN apt update && apt install -y libucx0 pipx python3-minimal python3-dev python3-pip python3-venv && \ RUN apt update && apt install -y libucx0 pipx python3-minimal python3-dev python3-pip python3-venv && \
rm -rf /var/lib/{apt,dpkg,cache,log}/ && \ rm -rf /var/lib/{apt,dpkg,cache,log}/ && \
pipx ensurepath && \ pipx ensurepath && \
@ -124,7 +126,7 @@ COPY --from=tgi-builder /usr/local/tgi /usr/local/tgi
COPY --from=tgi-builder /usr/src/text-generation-inference/target/release/text-generation-backends-trtllm /usr/local/tgi/bin/text-generation-launcher COPY --from=tgi-builder /usr/src/text-generation-inference/target/release/text-generation-backends-trtllm /usr/local/tgi/bin/text-generation-launcher
# This is used only for the CI/CD # This is used only for the CI/CD
FROM nvidia/cuda:12.8.0-cudnn-runtime-ubuntu24.04 AS ci-runtime FROM nvidia/cuda:${CUDA_BASE}-cudnn-runtime-ubuntu24.04 AS ci-runtime
RUN apt update && apt install -y libasan8 libubsan1 libucx0 pipx python3-minimal python3-dev python3-pip python3-venv && \ RUN apt update && apt install -y libasan8 libubsan1 libucx0 pipx python3-minimal python3-dev python3-pip python3-venv && \
rm -rf /var/lib/{apt,dpkg,cache,log}/ && \ rm -rf /var/lib/{apt,dpkg,cache,log}/ && \
pipx ensurepath && \ pipx ensurepath && \