Cohere Rerank 3.5
Cohere Rerank 3.5 is a reranking model from Cohere that reorders search results by semantic relevance with a 4096 token context window.
CohereSemantic reranking for retrieval pipelines
Cohere Rerank 3.5 is a reranking model developed by Cohere and released in December 2024. It is designed to take an initial set of search results and reorder them based on semantic relevance to a query, making it a post-retrieval step in retrieval-augmented generation (RAG) and search pipelines. The model accepts a query and a list of candidate documents, then returns a ranked list with relevance scores. Its context window supports up to 4096 tokens per input.
Rerank 3.5 is intended for use in applications where the quality of ranked results matters, such as enterprise search, document retrieval, and question-answering systems. Unlike generative models, it does not produce text output — it scores and orders existing content. The model is available as a first-party Cohere offering, meaning it is served directly by Cohere rather than through a third-party provider. It is well-suited for developers who need to improve the precision of retrieval results without rewriting their underlying search infrastructure.
What Cohere Rerank 3.5 supports
Semantic Reranking
Reorders a list of candidate documents by semantic relevance to a query, returning a scored and ranked list rather than generating new text.
RAG Pipeline Integration
Functions as a post-retrieval step in retrieval-augmented generation pipelines, improving result precision before passing documents to a generative model.
Document Relevance Scoring
Assigns a numerical relevance score to each candidate document relative to the input query, enabling downstream filtering or threshold-based selection.
Long Document Context
Supports a context window of up to 4096 tokens per input, accommodating moderately long documents or passages in a single reranking call.
Multilingual Support
Rerank 3.5 is designed to handle queries and documents across multiple languages, making it usable in multilingual search and retrieval applications.
Ready to build with Cohere Rerank 3.5?
Get Started FreeCommon questions about Cohere Rerank 3.5
What is the context window for Cohere Rerank 3.5?
Cohere Rerank 3.5 supports a context window of 4096 tokens per input document.
Does Cohere Rerank 3.5 generate text?
No. Rerank 3.5 is a reranking model, not a generative model. It takes a query and a list of candidate documents and returns relevance scores and a ranked ordering — it does not produce new text.
What is Cohere Rerank 3.5 best used for?
It is best used as a post-retrieval step in search and RAG pipelines, where it reorders an initial set of retrieved documents by semantic relevance to improve result quality before passing them to a generative model or displaying them to users.
Who provides Cohere Rerank 3.5?
Cohere Rerank 3.5 is a first-party model served directly by Cohere, the company that developed it. It was released in December 2024.
Is pricing information available for Cohere Rerank 3.5?
Pricing details are not included in the available metadata. You should consult Cohere's official documentation or pricing page for current cost information.
Explore similar models
CohereCohere Rerank 4 Fast
Cohere Rerank 4 Fast is a reranking model from Cohere designed to score and reorder search results with a 4096 token context window.
CohereCohere Rerank 4 Pro
Cohere Rerank 4 Pro is a reranking model from Cohere designed to improve search result relevance with a 4096-token context window.
CohereCohere Embed 4
Cohere Embed 4 is an embedding model from Cohere with a 128,000-token context window, released in April 2025.
NvidiaLlama Nemotron Rerank VL 1B v2
Llama Nemotron Rerank VL 1B v2 is a vision-language reranking model from Nvidia with a 10,240-token context window.
Voyage AIVoyage Rerank 2.5
Voyage Rerank 2.5 is a reranking model from Voyage AI designed to improve document retrieval relevance with a 32,000-token context window.
QwenQwen3 Reranker 0.6B
Qwen3 Reranker 0.6B is a lightweight reranking model from Qwen designed to reorder retrieved documents with a 32,768-token context window.
Start building with Cohere Rerank 3.5
No API keys required. Create AI-powered workflows with Cohere Rerank 3.5 in minutes — free.