Add API usage

README.md

@@ -21,7 +21,7 @@ tags:
 
 # KURE-Reranker-nano
 
-**KURE-Reranker-nano** is a lightweight, approximately **150M-parameter** Korean-English bilingual reranker. It jointly encodes a query and a candidate document and predicts a scalar relevance score.
+**KURE-Reranker-nano** is a lightweight, approximately **150M-parameter** Korean-English bilingual reranker. It jointly encodes a query and a candidate document and predicts a scalar relevance score. It is also served through the [KURE API](https://kure.kullm.ai/).
 
 ## Key Characteristics
 
@@ -112,6 +112,39 @@ for index in torch.argsort(scores, descending=True).tolist():
 
 </details>
 
+<details>
+<summary>API</summary>
+
+KURE-Reranker-nano is served at [kure.kullm.ai](https://kure.kullm.ai/). Sign in with GitHub there to get an API key.
+
+```python
+import os
+
+import requests
+
+query = "대한민국의 수도는 어디인가요?"
+documents = [
+    "대한민국의 수도는 서울입니다.",
+    "The capital of South Korea is Seoul.",
+    "바나나는 열대 지역에서 재배되는 과일입니다.",
+]
+
+# Results come sorted by relevance_score (sigmoid of the logit, 0 to 1). top_n keeps only the best N
+response = requests.post(
+    "https://kure.kullm.ai/api/v1/rerank",
+    headers={"Authorization": f"Bearer {os.environ['KURE_API_KEY']}"},
+    json={"model": "kure-reranker-nano", "query": query, "documents": documents, "return_documents": True},
+)
+response.raise_for_status()
+for result in response.json()["results"]:
+    print(f"{result['relevance_score']:.4f}\t{result['document']['text']}")
+# 0.9984	대한민국의 수도는 서울입니다.
+# 0.8661	The capital of South Korea is Seoul.
+# 0.0000	바나나는 열대 지역에서 재배되는 과일입니다.
+```
+
+</details>
+
 ## Korean Reranking Evaluation
 
 Scores are **nDCG@10**, and throughput is pairs per second (**PPS**).