perf: overlap Qwen shared expert with DeepEP routed experts (#34938)

docs/docs/references/environment_variables.mdx

@@ -331,6 +331,11 @@ SGLang supports various environment variables that can be used to configure its
       <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>Run shared experts on an alternate stream when single batch overlap is enabled on GB200. When not setting this flag, shared experts and down gemm will be overlapped with DeepEP combine together.</td>
       <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}>`"false"`</td>
     </tr>
+    <tr>
+      <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}>`SGLANG_ENABLE_QWEN_DEEPEP_SHARED_OVERLAP`</td>
+      <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>Run the Qwen shared expert on an alternate CUDA stream while DeepEP executes the routed experts.</td>
+      <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.02)"}}>`"true"`</td>
+    </tr>
     <tr>
       <td style={{padding: "9px 12px", fontWeight: 500, backgroundColor: "rgba(255,255,255,0.02)"}}><code>SGLANG_DISABLED_MODEL_ARCHS</code></td>
       <td style={{padding: "9px 12px", backgroundColor: "rgba(255,255,255,0.05)"}}>Comma-separated list of model architectures to disable from auto-registration.</td>