Dockerfile.cpu

# ===----------------------------------------------------------------------=== #
# Copyright (c) 2025, Modular Inc. All rights reserved.
#
# Licensed under the Apache License v2.0 with LLVM Exceptions:
# https://llvm.org/LICENSE.txt
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ===----------------------------------------------------------------------=== #

# This is an example of how to run max-pipelines serve in a docker container.

# Create a build image to download depedencies
FROM ghcr.io/modular/magic:latest AS production

RUN apt-get update && apt-get install -y --no-install-recommends build-essential git

COPY . /app
WORKDIR /app

RUN magic install

EXPOSE 8000

# TODO: Add HEALTHCHECK instruction
# HEALTHCHECK --interval=30s --timeout=30s --start-period=5s --retries=3 CMD curl --fail http://localhost:8000/health || exit 1

LABEL maintainer "Modular Inc <hello@modular.com>"

ENV PYTHONUNBUFFERED=1
# suppress transformers logging
ENV TRANSFORMERS_VERBOSITY="critical"
ENV TRANSFORMERS_NO_ADVISORY_WARNINGS=1
ENV MODULAR_STRUCTURED_LOGGING=1
ENV PROMETHEUS_MULTIPROC_DIR=/tmp

# OTEL configuration
ENV OTEL_SERVICE_NAME="max-serve"
# OTEL's default histogram bucktes have poor resolution.  Prefer exponential histograms.
ENV OTEL_EXPORTER_OTLP_METRICS_DEFAULT_HISTOGRAM_AGGREGATION="base2_exponential_bucket_histogram"

ENTRYPOINT ["magic", "run", "serve"]
CMD ["--huggingface-repo-id=modularai/Llama-3.1-8B-Instruct-GGUF"]