Model comparison
Compare Cohere Rerank 4 Fast and Voyage Rerank 3 using the same provider-sourced reranking & retrieval rubric. No mystery score and no invented benchmark ranking.
Facts checked September 4, 2026
Quick take
Cohere's lower-latency Rerank 4 option for high-volume search, recommendations, and RAG retrieval.
Fast is a relative product tier, not a latency guarantee. Measure it with your document lengths, region, candidate count, and traffic pattern.
Voyage AI's quality-focused preview reranker, billed by all query and candidate tokens processed in each request.
Treat it as a preview and confirm candidate caps, language behavior, retention, regional processing, and production support directly before launch.
Compare the published facts
Values use each provider's own published units and limits. A blank means the provider did not publish a directly comparable value in the sources reviewed.
| Reranking & retrieval | Cohere Rerank 4 Fast | Voyage Rerank 3 |
|---|---|---|
| Price basisThe provider's billing unit, with the document or token assumptions needed to compare it fairly. | $2 / 1K searches; up to 100 docs per unit | $0.05 / 1M processed tokens |
| Text capacityHow much query and candidate text the model can consider in one scoring pass. | 32,768 tokens per query-document pass | 32,000 tokens |
| Candidates per requestThe maximum number of possible results accepted in one request, when the provider publishes it. | Up to 10,000 documents per request | Not published for the Rerank 3 preview |
| Language coveragePublished multilingual support; an exact count is shown only when the provider gives one. | Multilingual; exact count not published | Not yet detailed for the preview |
| Structured dataWhether the model is documented for JSON, tables, XML, or other data beyond normal prose. | Text and semi-structured JSON | Text candidates; structured-data support not stated |
| Where it runsHosted API, cloud marketplace, private deployment, or self-hosted weights. | Cohere API; enterprise private and cloud options | Voyage hosted API and SDKs |
How to choose
Start with the job you need to complete, then validate cost, access, and policy details on your exact provider route.
Cohere enterprise customers can opt out of training; SaaS prompts and generations are generally deleted after 30 days. Approved zero-data-retention accounts and private or third-party deployments offer stronger controls.
Provider and API links
The reviewed Rerank 3 and pricing documentation does not publish a complete prompt-retention or model-training commitment. Obtain current data-processing terms before sending private retrieval candidates.
Provider and API links
Frequently asked questions
Cohere Rerank 4 Fast: Cohere's lower-latency Rerank 4 option for high-volume search, recommendations, and RAG retrieval. Voyage Rerank 3: Voyage AI's quality-focused preview reranker, billed by all query and candidate tokens processed in each request.
Consider Cohere Rerank 4 Fast when your priority is High-throughput product search. Consider Voyage Rerank 3 when your priority is Teams already using Voyage embeddings. Test both with your own data and provider route before committing.
No. This comparison aligns provider-published facts for the Reranking & retrieval category. It does not claim a universal winner or combine incompatible third-party benchmark scores.
