[Docs] Add LightLLM deployment example

README.md

@@ -156,6 +156,29 @@ python3 -m sglang.launch_server \
 ```
 + For Blackwell GPUs, include `--attention-backend triton --speculative-draft-attention-backend triton` in your SGLang launch command.
 
+### LightLLM
+```shell
+# Install lightllm (Recommended to use Docker).
+docker pull jyily/lightllm:cu129-78cc66a  
+
+# Or build from [source](https://github.com/ModelTC/LightLLM)
+pip install git+https://github.com/ModelTC/LightLLM
+
+LIGHTLLM_TRITON_AUTOTUNE_LEVEL=1 LOADWORKER=18 \
+python3 -m lightllm.server.api_server \
+    --model_dir /path/to/GLM-4.7-Flash/ \
+    --tp 1 \
+    --max_req_total_len 202752 \
+    --chunked_prefill_size 8192 \
+    --llm_prefill_att_backend fa3 \
+    --llm_decode_att_backend fa3 \
+    --graph_max_batch_size 512 \
+    --tool_call_parser glm47 \
+    --reasoning_parser glm45 \
+    --host 0.0.0.0 \
+    --port 8000
+```
+
 ## Citation
 
 If you find our work useful in your research, please consider citing the following paper: