Cohere: Cohere Rerank 4 Pro

Cohere Rerank 4 Pro is the quality-oriented multilingual reranking model in the Rerank 4 family. It scores retrieved text against a query for more demanding search and RAG workflows; billing uses reported search units rather than an estimated token count.

Compare models

rerank-v4.0-pro

AvailableCohereRerankSync

Price: From $0.0025
Modalities: Reranking

About Cohere Rerank 4 Pro

Cohere Rerank 4 Pro is the quality-oriented model in Cohere's Rerank 4 family. It scores retrieved text against a query for demanding search and RAG work, where Cohere says deeper reasoning and precision matter most. It reads multilingual text and semi-structured content such as JSON and tables, and offers the larger document window of the Rerank 4 generation.

Where it works well

Cohere positions Pro for the highest ranking quality and for complex use cases.
Deeper analysis of how a passage relates to a query, per Cohere's description of Pro.
Multilingual reranking across the languages Cohere lists for Rerank 4.
Semi-structured documents such as JSON, tables and code-like text are supported.
Longer documents and multiple passages can be judged together with the expanded window.

When to choose another model

Fast is the better pick when response time or volume dominates quality.
It orders candidates only; the answer text still comes from a generation model.
It cannot recover documents the first-stage retrieval missed.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

RerankingTokenLab endpoint

POST/v1/rerank

curl -X POST "https://api.tokenlab.sh/v1/rerank" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "operation": "rerank",
  "model": "rerank-v4.0-pro",
  "query": "What is the capital of France?",
  "documents": [
    "Paris is the capital of France.",
    "Berlin is the capital of Germany."
  ]
}'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Reranking

per request

Official price: $0.0025
Verified price: —
Discount: —

	Official priceper request	Verified priceper request	Discount
Reranking	$0.0025	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 14, 2026, 11:13 PM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Cohere Rerank 4 Pro in Console with a prompt ready to edit or send.

Help me try rerank-v4.0-pro with rerank at /v1/rerank. Show the result and cost.

Use cases

Enterprise knowledge search
Rerank candidates from many internal sources where wrong top results are costly.
Legal and policy retrieval
Surface the clause that answers a precise question from long, similar-sounding documents.
Technical support
Order troubleshooting articles for rare, specific error messages, where several near-identical pages compete for the top position.
Evaluation baselines
Use it as the quality reference when testing whether a lighter reranker is good enough for your own data.

Prompt examples

Query: which contract clause allows termination for convenience? Documents: twenty clause excerpts.

Query: error 0x80070057 during install. Documents: eight knowledge base entries.

Query: 保险理赔流程 Documents: ten Chinese and English FAQ entries.

FAQ

What is Rerank 4 Pro?

Rerank 4 Pro is Cohere's quality-optimized reranker. It takes a query and a list of documents and orders them by relevance, for use between retrieval and generation.

When should I choose Pro over Fast?

Choose Pro when ranking accuracy on hard or ambiguous queries is worth the extra processing time. Use Fast for latency-sensitive paths where an approximate order is acceptable.

Does Pro support long documents?

Cohere says Rerank 4 has a larger window than earlier generations, so longer passages and several sections fit in one judgment. That helps with long manuals and multi-part records.

Does it support multiple languages?

Yes, multilingual reranking is part of the Rerank 4 family, so queries and documents may be written in different languages. Test your own language pair on a sample set to confirm quality.

Does it return answers?

No. It returns relevance scores and an ordering of your documents. Pass the top results to a chat model if you want an answer written from them.

How much does Cohere Rerank 4 Pro cost?

On TokenLab, Cohere Rerank 4 Pro costs $0.0025 per request. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Cohere Rerank 4 Pro use?

Use https://api.tokenlab.sh/v1/rerank for Cohere Rerank 4 Pro. The request example below shows the matching code shape.

Which operations does Cohere Rerank 4 Pro support?

Cohere Rerank 4 Pro supports Reranking. Select an operation above to see its endpoint and request example.