Skip to content

Rerank 3.5

cohere/rerank-3.5
Search

The retrieval quality multiplier, reorders candidate documents by true relevance, per query.

Model overview

Context window

4K

tokens

Input price

quoted on request

Output price

quoted on request

Weekly volume

tokens / week

Open weights

No

API only

Capabilities

1

of 11 tags

Description

Rerank 3.5 is a search model from Cohere. Webparam does not stock this model yet and can source it on request — tell us what you need it for and we will come back with availability and a price. Cohere’s stated focus is enterprise retrieval and language stack. The weights are not published, so it is API-only.

Capabilities

Batch
Accepts offline batch jobs, which are processed asynchronously at a reduced rate.

Strengths & weaknesses

Strengths

  • Large RAG quality gains for minimal change
  • Fast per-query latency
  • Multilingual

Weaknesses

  • Per-query pricing needs its own budgeting
  • 4K window per document

Pricing

Free
Pricing for Rerank 3.5, per 1K queries
RatePriceUnit
Unit price$0.00per 1K queries

Batch jobs are supported and settle at a reduced rate against the same unit. Free at the point of use; fair-use rate limits apply to the free tier. Illustrative placeholder pricing.

Context window

4Ktokens

covers prompt and response together, so a long input leaves less room for the answer.

4K tokens against a peer range of 4K to 10M, median 200K. Compared across all models. This category has too few peers to scale against. Logarithmic scale.

Supported features

Feature support for Rerank 3.5
FeatureSupport
Tool callingNot supported
JSON modeNot supported
StreamingNot supported
Vision inputNot supported
Audio inputNot supported
Long contextNot supported
Open weightsNot supported
ReasoningNot supported
Fine-tunableNot supported
BatchSupported
CachingNot supported

Example use cases

  • RAG reranking

    Large RAG quality gains for minimal change

  • Search relevance

    Fast per-query latency

  • Recommendation ordering

    Multilingual

Comparison

Rerank 3.5 compared with Embed v4 and Command A
AttributeRerank 3.5Embed v4Command A
Context window4K128K256K
Input price$0.00$0.00$0.00
Output price not charged not charged not charged
Tool callingNot supportedNot supportedSupported
JSON modeNot supportedNot supportedSupported
Vision inputNot supportedSupportedNot supported
Open weightsNot supportedNot supportedNot supported
Prompt cachingNot supportedNot supportedNot supported

Best value in each row is highlighted. Illustrative placeholder data.

Provider information

View Cohere
Cohere3 models

Cohere · Toronto, Canada · est. 2019

Cohere never chased consumer chat. It built the enterprise retrieval stack instead.

Recent releases

    Documentation

    Frequently asked questions

    Usage is billed at $0.00 per 1K queries. Batch jobs settle at a reduced rate against the same unit. Every figure on this page is an illustrative placeholder.