Update README.md

README.md

@@ -81,7 +81,7 @@ docker run --rm --gpus all --ipc=host --shm-size 32g \
     --port 30000
 ```
 
-On 80GB cards (H100 / H800), scale out to `--tp 8`. The reasoning and tool-call parsers resolve automatically to `ling3` from the chat template; you can also set them explicitly with `--reasoning-parser ling3 --tool-call-parser ling3`.
+On 80GB cards (H100 / H800), scale out to `--tp 2`. The reasoning and tool-call parsers resolve automatically to `ling3` from the chat template; you can also set them explicitly with `--reasoning-parser ling3 --tool-call-parser ling3`.
 
 **Client**
 
@@ -127,7 +127,7 @@ vllm serve "$MODEL_PATH" \
     --port "$PORT" \
     --trust-remote-code \
     --served-model-name auto \
-    --tensor-parallel-size 4 \
+    --tensor-parallel-size 2 \
     --gpu-memory-utilization 0.85 \
     --enable-prefix-caching \
     --mamba-cache-mode align \