docs: add vllm 0.8 page (#694)

## What does this PR do?

Add document for using vLLM 0.8 in verl

## Who can review?

@eric-haibin-lin

README.md

@@ -108,8 +108,8 @@ verl is fast with:
 ## Performance Tuning Guide
 The performance is essential for on-policy RL algorithm. We write a detailed performance tuning guide to allow people tune the performance. See [here](https://verl.readthedocs.io/en/latest/perf/perf_tuning.html) for more details.
 
-## vLLM v0.7 integration preview
-We have released a testing version of veRL that supports vLLM>=0.7.0. Please refer to [this document](https://github.com/volcengine/verl/blob/main/docs/README_vllm0.7.md) for installation guide and more information.
+## Use vLLM v0.8
+veRL now supports vLLM>=0.8.0 when using FSDP as the training backend. Please refer to [this document](docs/README_vllm0.8.md) for installation guide and more information.
 
 ## Citation and acknowledgement
 

docker/Dockfile.ngc.vllm0.8

+# Start from the NVIDIA official image (ubuntu-22.04 + python-3.10)
+# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-24-08.html
+FROM nvcr.io/nvidia/pytorch:24.08-py3
+
+# uninstall nv-pytorch fork
+RUN pip3 uninstall -y pytorch-quantization \
+    pytorch-triton torch torch-tensorrt torchvision \
+    xgboost transformer_engine flash_attn apex megatron-core
+
+# Define environments
+ENV MAX_JOBS=32
+ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
+ENV DEBIAN_FRONTEND=noninteractive
+ENV NODE_OPTIONS=""
+ENV HF_HUB_ENABLE_HF_TRANSFER="1"
+
+# Define installation arguments
+ARG APT_SOURCE=https://mirrors.tuna.tsinghua.edu.cn/ubuntu/
+ARG PIP_INDEX=https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+
+# Set apt source
+RUN cp /etc/apt/sources.list /etc/apt/sources.list.bak && \
+    { \
+    echo "deb ${APT_SOURCE} jammy main restricted universe multiverse"; \
+    echo "deb ${APT_SOURCE} jammy-updates main restricted universe multiverse"; \
+    echo "deb ${APT_SOURCE} jammy-backports main restricted universe multiverse"; \
+    echo "deb ${APT_SOURCE} jammy-security main restricted universe multiverse"; \
+    } > /etc/apt/sources.list
+
+# Install systemctl
+RUN apt-get update && \
+    apt-get install -y -o Dpkg::Options::="--force-confdef" systemd && \
+    apt-get clean
+
+# Install tini
+RUN apt-get update && \
+    apt-get install -y tini && \
+    apt-get clean
+
+# Change pip source
+RUN pip config set global.index-url "${PIP_INDEX}" && \
+    pip config set global.extra-index-url "${PIP_INDEX}" && \
+    python -m pip install --upgrade pip
+
+# Install torch-2.6.0 + vllm-0.8.1
+RUN pip install --no-cache-dir vllm==0.8.1 torch==2.6.0 torchvision==0.21.0 torchaudio==2.6.0 tensordict torchdata \
+    transformers>=4.49.0 accelerate datasets peft hf-transfer \
+    ray codetiming hydra-core pandas pyarrow>=15.0.0 pylatexenc qwen-vl-utils wandb dill pybind11 liger-kernel mathruler \
+    pytest yapf py-spy pyext pre-commit ruff
+
+# Install flash_attn-2.7.4.post1
+RUN pip uninstall -y transformer-engine flash-attn && \
+    wget -nv https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.4.post1/flash_attn-2.7.4.post1+cu12torch2.6cxx11abiFALSE-cp310-cp310-linux_x86_64.whl && \
+    pip install --no-cache-dir flash_attn-2.7.4.post1+cu12torch2.6cxx11abiFALSE-cp310-cp310-linux_x86_64.whl
+
+# Fix cv2
+RUN pip uninstall -y pynvml nvidia-ml-py && \
+    pip install --no-cache-dir nvidia-ml-py>=12.560.30 opencv-python-headless==4.8.0.74 fastapi==0.115.6 && \
+    pip install -U optree>=0.13.0

docs/README_vllm0.8.md

+# Upgrading to vLLM >= 0.8
+
+## Installation
+
+Note: This version of veRL+vLLM 0.8+ supports **FSDP** for training and **vLLM** for rollout.
+
+```bash
+# Create the conda environment
+conda create -n verl python==3.10
+conda activate verl
+
+# Install verl
+git clone https://github.com/volcengine/verl.git
+cd verl
+pip3 install -e .
+
+# Install the latest stable version of vLLM
+pip3 install vllm==0.8.1
+
+# Install flash-attn
+pip3 install flash-attn --no-build-isolation
+
+```
+
+We have a pre-built docker image for veRL+vLLM 0.8.0. You can direct import it with the following command:
+
+```bash
+docker pull hiyouga/verl:ngc-th2.6.0-cu120-vllm0.8.0
+```
+
+## Features
+
+vLLM 0.8+ supports cuda graph and V1 engine by default in veRL. To enable these features, remember to add the following lines to the bash script:
+
+```bash
+actor_rollout_ref.rollout.enforce_eager=False \
+actor_rollout_ref.rollout.free_cache_engine=False \
+```
+
+and also **remove** the environment variable if it exists:
+
+```bash
+export VLLM_ATTENTION_BACKEND=XFORMERS
+```

docs/index.rst

@@ -74,6 +74,7 @@ verl is fast with:
    
    perf/perf_tuning
    README_vllm0.7.md
+   README_vllm0.8.md
 
 .. toctree::
    :maxdepth: 1