[OpenVINO] Updated OpenVINO requirements and build docs

Dockerfile.openvino

@@ -1,7 +1,7 @@
 # The vLLM Dockerfile is used to construct vLLM image that can be directly used
 # to run the OpenAI compatible server.
 
-FROM ubuntu:20.04 AS dev
+FROM ubuntu:22.04 AS dev
 
 RUN apt-get update -y && \
     apt-get install -y python3-pip git
@@ -18,7 +18,7 @@ COPY setup.py /workspace/vllm/
 # install build requirements
 RUN PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" python3 -m pip install -r /workspace/vllm/requirements-build.txt
 # build vLLM with OpenVINO backend
-RUN PIP_PRE=1 PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu https://storage.openvinotoolkit.org/simple/wheels/nightly/" VLLM_TARGET_DEVICE="openvino" python3 -m pip install /workspace/vllm/
+RUN PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu https://storage.openvinotoolkit.org/simple/wheels/pre-release" VLLM_TARGET_DEVICE="openvino" python3 -m pip install /workspace/vllm/
 
 COPY examples/ /workspace/vllm/examples
 COPY benchmarks/ /workspace/vllm/benchmarks

docs/source/getting_started/openvino-installation.rst

@@ -57,7 +57,7 @@ Install from source
 
   .. code-block:: console
 
-      $ PIP_PRE=1 PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu https://storage.openvinotoolkit.org/simple/wheels/nightly/" VLLM_TARGET_DEVICE=openvino python -m pip install -v .
+      $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu https://storage.openvinotoolkit.org/simple/wheels/pre-release" VLLM_TARGET_DEVICE=openvino python -m pip install -v .
 
 .. _openvino_backend_performance_tips:
 

requirements-openvino.txt

@@ -1,7 +1,33 @@
 # Common dependencies
--r requirements-common.txt
+# -r requirements-common.txt
+# TODO: remove temporary copy of all common dependencies once Optimum Intel will support Transformers >= 4.43.2
+cmake >= 3.21
+ninja  # For faster builds.
+psutil
+sentencepiece  # Required for LLaMA tokenizer.
+numpy < 2.0.0
+requests
+tqdm
+py-cpuinfo
+transformers < 4.43
+tokenizers >= 0.19.1  # Required for Llama 3.
+fastapi
+aiohttp
+openai
+uvicorn[standard]
+pydantic >= 2.0  # Required for OpenAI server.
+pillow  # Required for image processing
+prometheus_client >= 0.18.0
+prometheus-fastapi-instrumentator >= 7.0.0
+tiktoken >= 0.6.0  # Required for DBRX tokenizer
+lm-format-enforcer == 0.10.3
+outlines >= 0.0.43, < 0.1 # Requires torch >= 2.1.0
+typing_extensions
+filelock >= 3.10.4 # filelock starts to support `mode` argument from 3.10.4
+pyzmq
 
 # OpenVINO dependencies
 torch >= 2.1.2
 openvino ~= 2024.3.0.dev
+openvino-tokenizers[transformers] ~= 2024.3.0.0.dev
 optimum-intel[openvino] >= 1.18.1