[docs] Add gpt oss deployment example

doc/source/serve/tutorials/deployment-serve-llm/README.ipynb

@@ -39,7 +39,12 @@
     "---\n",
     "\n",
     "**[Deploy a hybrid reasoning LLM](https://docs.ray.io/en/latest/serve/tutorials/deployment-serve-llm/hybrid-reasoning-llm/README.html)**  \n",
-    "Deploy models that can switch between reasoning and non-reasoning modes for flexible usage, such as Qwen-3."
+    "Deploy models that can switch between reasoning and non-reasoning modes for flexible usage, such as Qwen-3.\n",
+    "\n",
+    "---\n",
+    "\n",
+    "**[Deploy gpt-oss](https://docs.ray.io/en/latest/ray-overview/examples/deployment-serve-llm/gpt-oss/README.html)**  \n",
+    "Deploy gpt-oss reasoning models, including `gpt-oss-20b` for lower latency use cases and `gpt-oss-120b` for high-reasoning, production-scale workloads."
    ]
   }
  ],

doc/source/serve/tutorials/deployment-serve-llm/README.md

@@ -39,3 +39,8 @@ Deploy models with reasoning capabilities designed for long-context tasks, codin
 
 **[Deploy a hybrid reasoning LLM](https://docs.ray.io/en/latest/serve/tutorials/deployment-serve-llm/hybrid-reasoning-llm/README.html)**  
 Deploy models that can switch between reasoning and non-reasoning modes for flexible usage, such as Qwen-3.
+
+---
+
+**[Deploy gpt-oss](https://docs.ray.io/en/latest/ray-overview/examples/deployment-serve-llm/gpt-oss/README.html)**  
+Deploy gpt-oss reasoning models, including `gpt-oss-20b` for lower latency use cases and `gpt-oss-120b` for high-reasoning, production-scale workloads.

doc/source/serve/tutorials/deployment-serve-llm/ci/tests.sh

@@ -12,7 +12,8 @@ for nb in \
   "large-size-llm/notebook" \
   "vision-llm/notebook" \
   "reasoning-llm/notebook" \
-  "hybrid-reasoning-llm/notebook"
+  "hybrid-reasoning-llm/notebook" \
+  "gpt-oss/notebook" 
 do
   python ci/nb2py.py "${nb}.ipynb" "${nb}.py" --ignore-cmds
   python "${nb}.py"

doc/source/serve/tutorials/deployment-serve-llm/gpt-oss/Dockerfile

@@ -0,0 +1,8 @@
+FROM anyscale/ray:2.49.0-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==0.10.1