Update README.md

README.md

@@ -32,6 +32,10 @@ label_names = ["entailment", "neutral", "contradiction"]
 prediction = {name: round(float(pred) * 100, 1) for pred, name in zip(prediction, label_names)}
 print(prediction)
 ```
+
+Example cURL:
+
+''''curl YOUR INFERENCE ENDPOINT URL HERE \ -X POST \ -d '{"inputs": {"premise": "A man is walking his dog in the park.", "hypothesis": "A person is outside with an animal."}}' \ -H "Authorization: Bearer hf_YOUR_TOKEN_HERE" \ -H "Content-Type: application/json
 ### Training data
 This model was trained on the MultiNLI dataset, which consists of 392 702 NLI hypothesis-premise pairs.