# Rerank 3.5

> Cohere's Rerank 3.5 scores how relevant each document is to a query. It handles 100+ languages and long, semi-structured documents (emails, tables, JSON, code). Use it as a second stage after vector or keyword search to put the best passages first.

**ID**: `cohere/rerank-v3.5`  
**Creator**: Cohere  
**Category**: rerank  
**Context**: 4K tokens  
**Input**: $2.00/M  
**Output**: $0  
**Released**: 2024-12-02  
**Web page**: https://anyrouter.dev/model/cohere/rerank-v3.5

**Input modalities**: text  
**Output modalities**: scores  

**Capabilities**: rerank

## Usage

**Endpoint**: `POST https://anyrouter.dev/api/v1/rerank`

```bash
curl https://anyrouter.dev/api/v1/rerank \
  -H "Authorization: Bearer $ANYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "cohere/rerank-v3.5",
  "query": "What is the capital of France?",
  "documents": [
    "Paris is the capital and largest city of France.",
    "Berlin is the capital of Germany.",
    "Bananas are rich in potassium."
  ],
  "top_n": 2
}'
```

Rerank is not chat: send a `query` and `documents`; the response lists `results` sorted by `relevance_score`.

## Providers

| Provider | Input | Output |
| --- | --- | --- |
| cohere-rerank | $2.00/M | $0 |
| cohere-rerank-byok | $0 | $0 |

## Source

- Upstream docs: https://docs.cohere.com/docs/rerank