cohere-rerank-4-fast - Pinecone Docs

When to use it

Cohere Rerank 4.0 Fast (cohere-rerank-4-fast) is the latest iteration in Cohere’s Rerank model series, succeeding Cohere Rerank v3.5. It improves relevance quality on queries that express explicit or implicit constraints, and it retains multilingual support for 100+ languages with strong performance in domains like finance and hospitality.

cohere-rerank-4-fast replaces cohere-rerank-3.5, which is deprecated. Since August 31, 2026, requests to cohere-rerank-3.5 are served by cohere-rerank-4-fast. Because the two models return different relevance scores, update your rerank requests to specify cohere-rerank-4-fast and re-tune any hard-coded score thresholds against it.

Inputs, limits, and pricing

Input	Limits	Price
Text	4096 tokens per document, 250 documents per request	$2.00 per 1k rerank units

Monthly and per-minute limits depend on your plan. See Rate limits.

Quickstart

Installation

pip install -U pinecone

Rerank

See rerank to learn more about reranking.

from pinecone import Pinecone

pc = Pinecone("API-KEY")

query = "Tell me about Apple's products"
results = pc.inference.rerank(
    model="cohere-rerank-4-fast",
    query=query,
    documents=[
"Apple is a popular fruit known for its sweetness and crisp texture.",
"Apple is known for its innovative products like the iPhone.",
"Many people enjoy eating apples as a healthy snack.",
"Apple Inc. has revolutionized the tech industry with its sleek designs and user-friendly interfaces.",
"An apple a day keeps the doctor away, as the saying goes.",
    ],
    top_n=3,
    return_documents=True
)

print(query)
for r in results.data:
  print(r.score, r.document["text"])

Parameters

Set rank_fields on the rerank request itself, and set the other parameters in its parameters object. This applies to both standalone reranking and integrated reranking.

Parameter	Type	Required/Optional	Description	Default
`max_tokens_per_doc`	integer	Optional	Documents longer than this are truncated. Max `4096`.	`1024`
`rank_fields`	array of strings	Optional	The fields to use for reranking, in priority order (e.g., `["field1", "field2"]`).	`["text"]`

Good to know

Cohere Rerank 4.0 is hosted on Azure AI under the Global Standard deployment type, and requests may be processed in regions outside the United States.Reranking with cohere-rerank-4-fast is billed per rerank unit. Most requests count as a single unit. One rerank unit covers a query plus up to 100 documents, and documents longer than about 500 tokens are split into ~500-token chunks that each count toward that 100, so a request with more than 100 chunks costs more than one unit. Before the transition, cohere-rerank-3.5 billed one unit per request. To keep cost and latency predictable, use max_tokens_per_doc to truncate long documents. For details, see Understanding cost.