Update README.md

README.md

@@ -17,7 +17,7 @@ base_model_relation: quantized
 Base model: [ByteDance-Seed/Seed-OSS-36B-Instruct](https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct)
 
 ### 【❗❗Reminder❗❗】
-The model suffers from significant loss under 3-bit quantization, please use with caution.
+Int3 quantization may cause measurable degradation in model accuracy. We recommend exercising caution when using this model.
 
 ### 【vLLM Single Node with 2 GPUs — Startup Command】
 ```