Add instructions

README.md

@@ -2,19 +2,46 @@
 
 This repo is a fork of the [vLLM](https://github.com/vllm-project/vllm) repo.
 
-## Build the image
+## Usage
+
+Pull the `latest` image from ECR:
+
+```bash
+bash docker/pull.sh vllm:latest
+```
+
+Run the container (with Command-R model in this case):
+
+```bash
+docker run --runtime nvidia --gpus all \
+    -v ~/.cache/huggingface:/root/.cache/huggingface \
+    -p 8000:8000 \
+    --ipc=host \
+    -e SERVED_MODEL_NAME=command-r \
+    -e TRUST_REMOTE_CODE=false \
+    -e MODEL=CohereForAI/c4ai-command-r-v01 \
+    vllm \
+    --tensor-parallel-size 4 \
+    --host 0.0.0.0
+```
+
+## Development
+
+### Build the image
 
 ```bash
 sh docker/build.sh
 ```
 
-## Deploy the image to ECR
+### Deploy the image to ECR
+
+Once your changes are ready, you can deploy the image to ECR:
 
 ```bash
 sh docker/deploy.sh
 ```
 
-## Upgrade version
+### Upgrade version
 
 You can upgrade the version of vLLM by rebasing on the official repo: