Microsoft AI's high-volume, low-latency variant of MAI-Voice-2.1 for speech in 23 languages, aimed at voice agents
Cohere Rerank 5 Pro
New· 4Cohere's higher-accuracy reranking model for search and retrieval, released 21 October 2026 at $0.05 per million tokens, with a Context Focus option that returns only the relevant passages of each document.
TrustList summary
Not yet independently verified. Rests on Cohere's own announcement of 21 October 2026. The benchmark results are Cohere's own, measured with a metric Cohere developed, and no independent dated test was found; the page states no context length or licence. We will update this when it can be confirmed, and remove this note.
Cohere Rerank 5 Pro is a reranking model that Cohere released on 21 October 2026 as part of its latest generation of search models, alongside the faster Rerank 5 Fast. A reranker takes a query and a list of candidate documents found by a first search, and reorders them by how relevant each one is, so that the best evidence reaches a language model or a user first.
What it does
Cohere says Rerank 5 returns calibrated relevance scores, so that a score means roughly the same thing from one query to the next and can be used as a cut-off. The new Context Focus option goes further than ranking whole documents: it returns the most relevant passages inside each document, which Cohere says cuts the tokens passed on per task. Cohere presents the model as built for agentic workflows, where multi-step queries need evidence that may not closely resemble the original question.
On Cohere's own evaluation, Rerank 5 Pro ranked first on 39 of 46 composite datasets and at least second on 43, and scored 82.2 on BRIGHT, a reasoning-intensive retrieval benchmark, which Cohere says is 12.4 points above Rerank 4 Pro. Cohere measured these with RCP-nDCG@10, a metric it developed to deal with incomplete relevance judgements, so they cannot be compared directly with scores reported on standard nDCG.
Who it is for
The model suits teams running retrieval-augmented generation or enterprise search who already have a first-stage search and want better ordering before results go to a model. It is generally available on the Cohere API and Model Vault, Microsoft Foundry and Amazon SageMaker, to users of Cohere's North platform, and for private VPC or on-premises deployment.
Key facts
Maker: Cohere Released: 21 October 2026 Price: Pro $0.05, Fast $0.02 per 1M tokens on the Cohere API Available on: Cohere API, Model Vault, Microsoft Foundry, Amazon SageMaker, North Licence: not stated on the announcement page Variant: Pro, the higher-accuracy tier
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| No benchmark scores yet. Be the first to add one. | ||
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
Cohere Rerank 5 Pro launches
Trust Score
An earned signal from verification, reviews, awards, transparency and engagement - the vendor can't buy it.
Updated 10/28/2026
Request a demo or quote from Cohere Rerank 5 Pro
Reviews
Write the first review of Cohere Rerank 5 Pro
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about Cohere Rerank 5 Pro.
Cohere Rerank 5 Pro alternatives
Similar listings buyers compare against this one.
Open-source 2B decision model from AWS's Strands Agents project that picks options and returns scores with a reliability figure
Black Forest Labs' image generation and editing model with bounding-box layouts, multiple reference images and 4K output
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Cantina Security's first open-weight security model, a post-trained GLM-5.3-Flash for vulnerability research and exploit checks
Anthropic's smallest and fastest Claude 5.5 model, available on its own API and three cloud platforms
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
Ai2's open-weight 8B model that writes cited scientific reports from a research question and retrieved literature
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.