Update model card for Safetensors and vLLM

README.md

@@ -1,7 +1,39 @@
 ---
 language: en
 tags:
-- mlx
-library_name: mlx
+- safetensors
+- vllm
+- insurance
+- reinsurance
+library_name: transformers
 pipeline_tag: text-generation
 ---
+
+# Reinsure-8B (vLLM Compatible)
+
+This is a fine-tuned version of Meta's Llama 3.1 8B Instruct model, specifically trained on over 14,000 synthetic and real-world instruction pairs covering Reinsurance and Insurance domains (Catastrophe Modeling, Treaty Structures, Facultative documents, and regulatory compliance).
+
+## Format 
+This model has been successfully dequantized and exported to **16-bit Safetensors** format. It is fully compatible with any standard cloud inference engine, including **vLLM** and HuggingFace Text Generation Inference (TGI).
+
+## Deployment
+
+This model is intended for deployment on serverless GPU architectures (like **Modal**) using `vLLM` to act as a highly scalable context engine for an API gateway.
+
+### Example vLLM Deployment (Modal)
+```python
+import modal
+
+vllm_image = modal.Image.debian_slim().pip_install("vllm")
+app = modal.App("reinsure-8b")
+
+@app.function(image=vllm_image, gpu="L4", container_idle_timeout=300)
+@modal.web_server(8000)
+def serve():
+    import sys, subprocess
+    subprocess.Popen([
+        sys.executable, "-m", "vllm.entrypoints.openai.api_server",
+        "--model", "Reinsured-AI/Reinsure-8B",
+        "--port", "8000"
+    ])
+```