rerank-2.5-lite

About this model

Rerankers are neural networks that predict the relevancy scores between a query and documents and rank them based on the scores. They are used to refine search results in semantic search/retrieval systems and retrieval-augmented generation (RAG).

rerank-2.5-lite is a reranker optimized for both latency and quality, delivering a 7.16% improvement in retrieval accuracy over Cohere Rerank v3.5 across 93 datasets. It also outperformed Cohere Rerank v3.5 by 10.36% on the Massive Instructed Retrieval Benchmark (MAIR). The model supports a combined context length of 32K tokens per query–document pair, including up to 8K tokens for the query, enabling more accurate retrieval over longer documents. Additionally, rerank-2.5-lite supports instruction following, allowing users to guide relevance scoring through natural language prompts. Learn more about rerank-2.5-lite here: https://blog.voyageai.com/2025/08/11/rerank-2-5

Key model capabilities

• Optimized for both latency and quality, delivering a 7.16% improvement in retrieval accuracy over Cohere Rerank v3.5 across 93 datasets.
• Supports a combined context length of 32K tokens per query–document pair, including up to 8K tokens for the query, enabling more accurate retrieval over longer documents.
• Supports instruction following, allowing users to guide relevance scoring through natural language prompts.

Usage

The deployed Azure AI Foundry endpoint exposes the Voyage inference API. Authenticate with your Azure ML endpoint key or a bearer token issued for the workspace.

Rerank Documents

curl <AZUREML_ENDPOINT_URL>/v1/rerank \
  -X POST \
  -H "Authorization: Bearer <AZUREML_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What is vector search?",
    "documents": [
      "Vector search enables semantic similarity search using embeddings.",
      "SQL databases use row-based storage for structured data."
    ],
    "model": "rerank-2.5-lite"
  }'

Supported Parameters

• query (string, required): The search query.
• documents (array of strings, required): List of documents to rerank (up to 1,000).
• model (string, required): rerank-2.5-lite.
• top_k (int, optional): Return only the top K results. Defaults to returning all documents.
• return_documents (bool, optional): Include the document text in the response. Defaults to false.
• truncation (bool, optional): Truncate inputs exceeding the context limit. Defaults to true.

See the full API reference at https://docs.voyageai.com/reference/reranker-api .

Response

{
  "object": "list",
  "data": [
    {"relevance_score": 0.8984375, "index": 0},
    {"relevance_score": 0.123046875, "index": 1}
  ],
  "model": "rerank-2.5-lite",
  "usage": {"total_tokens": 42}
}

Each item in data contains a relevance_score (higher is more relevant) and the original index of the document. Results are returned sorted by descending relevance score.