Update README.md

README.md

@@ -1,20 +1,21 @@
 ---
-language:
-- en
-- zh
 tags:
 - fp8
-- quantization
-- dynamic
-- vision-language
-- multimodal
+- fp8-dynamic
 - vllm
 - llm-compressor
 - internvl3.5
-
+- internvl
+language:
+- multilingual
 pipeline_tag: image-text-to-text
 inference: false
 license: mit
+base_model:
+- OpenGVLab/InternVL3_5-38B
+datasets:
+- OpenGVLab/MMPR-v1.2
+library_name: vllm
 ---
 
 # InternVL3.5 38B FP8
@@ -41,7 +42,22 @@ The quantization process uses a specialized recipe that preserves the model's co
 | **Quantization Library** | [LLM Compressor](https://github.com/vllm-project/llm-compressor) v0.7.1 |
 | **Quantized By** | [brandonbeiler](https://huggingface.co/brandonbeiler) |
 
-## Usage with vLLM
+## With vLLM OpenAI-Compatible Server
+
+You can serve the model using vLLM's OpenAI-compatible API server.
+
+```bash
+python -m vllm.entrypoints.openai.api_server \
+    --model brandonbeiler/InternVL3_5-38B-FP8-Dynamic \
+    --quantization compressed-tensors \
+    --served-model-name internvl3_5-38b \
+    --reasoning-parser: qwen3 \
+    --trust-remote-code \
+    --max-model-len 32768 \
+    --tensor-parallel-size 1 # Adjust based on your GPU setup
+```
+
+## Usage with vLLM in Python
 
 The following snippet demonstrates inference using the vLLM library.
 
@@ -69,6 +85,8 @@ response = model.generate(prompt, sampling_params)
 print(response[0].outputs[0].text)
 ```
 
+
+
 ## Technical Specifications
 
 ### Hardware Requirements