API Reference
Rerank
Improve search and retrieval quality by reranking document relevance using models from Cohere, Jina AI, and other providers through a single endpoint.
Endpoint
http
POST https://api.onerouter.app/v1/rerankRequest Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | Yes | Rerank model ID (e.g., cohere-rerank-v3, jina-reranker-v2) |
| query | string | Yes | The search query to evaluate documents against |
| documents | array | Yes | List of document strings or objects to rerank |
| top_n | integer | No | Return only the top N results. Default: returns all |
| return_documents | boolean | No | Include document text in response. Default: true |
Example
Send a query with a list of candidate documents and receive relevance-scored results:
python
import requests
response = requests.post(
"https://api.onerouter.app/v1/rerank",
headers={
"Authorization": "Bearer sk-your-key",
"Content-Type": "application/json",
},
json={
"model": "cohere-rerank-v3",
"query": "What is the capital of France?",
"documents": [
"Paris is the capital of France.",
"Berlin is the capital of Germany.",
"France is a country in Europe.",
],
"top_n": 2,
},
)
for result in response.json()["results"]:
print(f"Score: {result['relevance_score']:.3f} — {result['document']['text']}")Common Use Cases
- RAG pipelines — Rerank retrieved chunks before passing to the LLM for more accurate answers
- Hybrid search — Combine keyword (BM25) and vector search results, then rerank for best relevance
- Multi-stage retrieval — Coarse retrieval → rerank → generation pipeline
Available Models
Supported reranking models include Cohere Rerank v3, Jina Reranker v2, and BGE-Reranker. See the Models page for the full catalog and per-model pricing.