Cohere Rerank AI features and pricing (October 2026)

Cohere Rerank is an API that takes a query and a list of documents and returns them sorted by relevance. Teams place it between their search step and the language model to improve the passages a RAG system uses.

Cohere, founded in 2019 and headquartered in Toronto, builds language models and enterprise AI products. Its rerank family currently includes rerank-v4.0-pro and rerank-v4.0-fast, plus the earlier rerank-v3.5 and version 3.0 models for English and multilingual text with a 4,096-token context. The models are called through the Rerank API endpoint. Cohere offers deployment through its own API, a managed Model Vault, private VPC setups and on-premises installations, and lists SOC 2 Type II, ISO 27001 and ISO 42001 on its security page.

• Improves search results before generationAfter a keyword or vector search returns candidates, the model scores each one against the query and reorders them. The language model then receives the most relevant passages, which reduces off-topic answers.
• Ranks documents in many languagesThe multilingual models handle English and non-English documents in one call. International teams can search mixed-language content without separate pipelines per language.
• Scores structured recordsRerank v4 models accept semi-structured JSON, such as product records or emails with fields. Teams can rank records by meaning, not only by exact field matches.

• Offers a pro and a fast variant of the current model, so teams can trade accuracy against latency.
• Available as a public API, as dedicated instances in Model Vault, in private VPCs and on premises.
• Cohere lists ISO 42001, a management system standard for AI, alongside SOC 2 Type II and ISO 27001.

Cohere is headquartered in Toronto, with further offices in Canada, the US, Europe and Korea.

Per-search prices for the standard API were not shown on the pricing page during our check in September 2026, so get a quote for your expected volume. The trial key is rate limited and not meant for production. Clarify where queries and documents are processed and whether the retention settings and training opt-out fit your data.

No reviews yet. Be the first to share how your team uses it.

Every review shows its verification level. We read each one before it goes live. How we check reviews