Rerank v4.0 Fast - MixRoute API

Cohere latency-focused multilingual relevance ranking model.

Model Information

Field	Description
`model`	rerank-v4.0-fast
`context`	32,768 tokens
`endpoint`	POST /v1/rerank

Context and output sizes describe the model specification. Effective request limits, availability, and billing depend on the MixRoute route.

Capabilities and Usage

Reranks supplied documents against a query; it does not generate answers or embeddings.
The context limit applies to a query/document pair. Retrieve candidates first, then use top_n to select the most relevant results.
results[].index maps back to the original documents array. Use that index instead of assuming document text is returned.

Request Example

Set MIXROUTE_API_KEY before making the request.

curl --request POST "https://api.mixroute.ai/v1/rerank" \
  --header "Authorization: Bearer $MIXROUTE_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "rerank-v4.0-fast",
  "query": "API documentation",
  "documents": [
    "An API reference for developers.",
    "A recipe for soup."
  ],
  "top_n": 1
}'

Request Fields

Field	Description
`model`	Required. Use the exact model ID above.
`query`	Required query text.
`documents`	Required array of document strings.
`top_n`	Optional number of results to return; no more than the number of candidates.

Complete request format: rerank.