diff --git a/deploy_deepseek_r1/Dockerfile b/deploy_deepseek_r1/Dockerfile
new file mode 100644
index 0000000..ea0023c
--- /dev/null
+++ b/deploy_deepseek_r1/Dockerfile
@@ -0,0 +1,8 @@
+FROM anyscale/ray:2.50.0-slim-py312-cu128
+
+# C compiler for Triton’s runtime build step (vLLM V1 engine)
+# https://github.com/vllm-project/vllm/issues/2997
+RUN sudo apt-get update && \
+    sudo apt-get install -y --no-install-recommends build-essential
+
+RUN pip install vllm==0.11.0
diff --git a/deploy_deepseek_r1/README.md b/deploy_deepseek_r1/README.md
new file mode 100644
index 0000000..db4036f
--- /dev/null
+++ b/deploy_deepseek_r1/README.md
@@ -0,0 +1,60 @@
+# Deploy DeepSeek R1
+
+This example uses Ray Serve along with vLLM to deploy a DeepSeek R1 model as an Anyscale service.
+
+## Install the Anyscale CLI
+
+```bash
+pip install -U anyscale
+anyscale login
+```
+
+## Deploy the service
+
+Clone the example from GitHub.
+
+```bash
+git clone https://github.com/anyscale/examples.git
+cd examples/deploy_deepseek_r1
+```
+
+Deploy the service. 
+```bash
+anyscale service deploy -f service.yaml
+```
+
+## Understanding the example
+
+- The [application code](https://github.com/anyscale/examples/blob/main/deploy_deepseek_r1/serve_deepseek_r1.py) sets the required accelerator type with `accelerator_type="H100"`. To use a different accelerator, replace `"H100"` with the desired name. See the [list of supported accelerators](https://docs.ray.io/en/latest/ray-core/accelerator-types.html#accelerator-types) for available options.
+- Ray Serve automatically autoscales the number of model replicas between `min_replicas` and `max_replicas`. Ray Serve adapts the number of replicas by monitoring queue sizes. For more information on configuring autoscaling, see the [AutoscalingConfig documentation](https://docs.ray.io/en/latest/serve/api/doc/ray.serve.config.AutoscalingConfig.html).
+- This example uses vLLM, and the [Dockerfile](https://github.com/anyscale/examples/blob/main/deploy_deepseek_r1/Dockerfile) defines the service’s dependencies. When you run `anyscale service deploy`, the build process adds these dependencies on top of an Anyscale-provided base image.
+- To configure vLLM, modify the `engine_kwargs` dictionary. See [Ray documentation for the `LLMConfig` object](https://docs.ray.io/en/latest/serve/api/doc/ray.serve.llm.LLMConfig.html#ray.serve.llm.LLMConfig).
+
+
+## Query the service
+
+The `anyscale service deploy` command outputs a line that looks like  
+```text
+curl -H "Authorization: Bearer <SERVICE_TOKEN>" <BASE_URL>
+```
+
+From the output, you can extract the service token and base URL. Open [query.py](https://github.com/anyscale/examples/blob/main/deploy_deepseek_r1/query.py) and add them to the appropriate fields.
+```python
+token = <SERVICE_TOKEN> 
+base_url = <BASE_URL> 
+```
+
+Query the model  
+```bash
+pip install openai
+python query.py
+```
+
+View the service in the [services tab](https://console.anyscale.com/services) of the Anyscale console.
+
+## Shutdown 
+ 
+Shutdown your Anyscale Service:
+```bash
+anyscale service terminate -n deploy-deepseek-r1
+```
\ No newline at end of file
diff --git a/deploy_deepseek_r1/query.py b/deploy_deepseek_r1/query.py
new file mode 100644
index 0000000..2034e60
--- /dev/null
+++ b/deploy_deepseek_r1/query.py
@@ -0,0 +1,33 @@
+from urllib.parse import urljoin
+from openai import OpenAI
+
+# The "anyscale service deploy" script outputs a line that looks like
+# 
+#     curl -H "Authorization: Bearer <SERVICE_TOKEN>" <BASE_URL>
+# 
+# From this, you can parse out the service token and base URL.
+token = <SERVICE_TOKEN>  # Fill this in. If deploying and querying locally, use token = "FAKE_KEY"
+base_url = <BASE_URL>  # Fill this in. If deploying and querying locally, use base_url = "http://localhost:8000"
+
+client = OpenAI(base_url= urljoin(base_url, "v1"), api_key=token)
+
+response = client.chat.completions.create(
+    model="my-deepseek-r1",
+    messages=[
+        {"role": "user", "content": "What's the capital of France?"}
+    ],
+    stream=True
+)
+
+# Stream and print JSON
+for chunk in response:
+    # Stream reasoning content
+    if hasattr(chunk.choices[0].delta, "reasoning_content"):
+        data_reasoning = chunk.choices[0].delta.reasoning_content
+        if data_reasoning:
+            print(data_reasoning, end="", flush=True)
+    # Later, stream the final answer
+    if hasattr(chunk.choices[0].delta, "content"):
+        data_content = chunk.choices[0].delta.content
+        if data_content:
+            print(data_content, end="", flush=True)
\ No newline at end of file
diff --git a/deploy_deepseek_r1/serve_deepseek_r1.py b/deploy_deepseek_r1/serve_deepseek_r1.py
new file mode 100644
index 0000000..a521467
--- /dev/null
+++ b/deploy_deepseek_r1/serve_deepseek_r1.py
@@ -0,0 +1,29 @@
+# serve_deepseek_r1.py
+from ray import serve
+from ray.serve.llm import LLMConfig, build_openai_app
+
+llm_config = LLMConfig(
+    model_loading_config=dict(
+        model_id="my-deepseek-r1",
+        model_source="deepseek-ai/DeepSeek-R1",
+    ),
+    accelerator_type="H100",
+    deployment_config=dict(
+        autoscaling_config=dict(
+            min_replicas=1,
+            max_replicas=1,
+        )
+    ),
+    engine_kwargs=dict(
+        max_model_len=16384,
+        # Split weights among 8 GPUs in the node
+        tensor_parallel_size=8,
+        pipeline_parallel_size=2,
+        reasoning_parser="deepseek_r1",  # Optional: separate reasoning content from the final answer
+    ),
+)
+
+app = build_openai_app({"llm_configs": [llm_config]})
+
+# Uncomment the below line to run the service locally with Python.
+# serve.run(app, blocking=True)
diff --git a/deploy_deepseek_r1/service.yaml b/deploy_deepseek_r1/service.yaml
new file mode 100644
index 0000000..edaf67c
--- /dev/null
+++ b/deploy_deepseek_r1/service.yaml
@@ -0,0 +1,59 @@
+# View the docs https://docs.anyscale.com/reference/service-api#serviceconfig.
+
+name: deploy-deepseek-r1
+
+# When empty, use the default image. This can be an Anyscale-provided base image
+# like anyscale/ray:2.43.0-slim-py312-cu125, a user-provided base image (provided
+# that it meets certain specs), or you can build new images using the Anyscale
+# image builder at https://console.anyscale-staging.com/v2/container-images.
+
+containerfile: ./Dockerfile
+
+# When empty, Anyscale will auto-select the instance types. You can also specify
+# minimum and maximum resources.
+compute_config:
+#   head_node:
+#     instance_type: m5.2xlarge
+#   worker_nodes:
+#     - instance_type: m5.16xlarge
+#       min_nodes: 0
+#       max_nodes: 100
+#     - instance_type: m7a.24xlarge
+#       min_nodes: 0
+#       max_nodes: 100
+#       market_type: PREFER_SPOT # Defaults to ON_DEMAND
+#     - instance_type: g4dn.2xlarge
+#       min_nodes: 0
+#       max_nodes: 100
+#       market_type: PREFER_SPOT # Defaults to ON_DEMAND
+  auto_select_worker_config: true
+  # Change default disk size to 1000GB
+  advanced_instance_config:
+    ## AWS ##
+    BlockDeviceMappings:
+      - Ebs:
+        - VolumeSize: 1000
+          VolumeType: gp3
+          DeleteOnTermination: true
+        DeviceName: "/dev/sda1"
+    #########
+    ## GCP ##
+    #instanceProperties:
+    #  disks:
+    #    - boot: true
+    #      auto_delete: true
+    #      initialize_params:
+    #        - disk_size_gb: 1000
+    #########
+
+# Path to a local directory or a remote URI to a .zip file (S3, GS, HTTP) that
+# will be the working directory for the job. The files in the directory will be
+# automatically uploaded to the job environment in Anyscale.
+working_dir: .
+
+# When empty, this uses the default Anyscale Cloud in your organization.
+cloud:
+
+# Specify the Ray Serve app to deploy.
+applications:
+- import_path: serve_deepseek_r1:app
\ No newline at end of file