Shared or dedicated inference for Embed & Rerank
Official developer update from Cohere (Cohere Blog).
Why it matters: Developer tooling release — update for the latest fixes, features, and compatibility.
All labs
Coverage & source status· 3/3 sources OK · checked Show
Community mirror of cohere.com/blog + CohereLabs on Hugging Face + GitHub.
- Cohere BlogRSScommunity mirrorsucceeded
- Hugging Face · CohereLabsHugging Facesucceeded
- GitHub · cohere-ai/cohere-pythonGitHub Releasessucceeded
36 updates · page 1 of 3
Friday, October 9, 2026
Developer Thursday, October 8, 2026
Feature Cohere Expands Canadian Footprint with New Ottawa Office
Cohere is expanding its Canadian footprint with a new office in Ottawa.
Why it matters: Customers and candidates in Canada can reach Cohere from the new Ottawa office.
Tuesday, October 6, 2026
Model Multilingual Bridges: How Data Mixing Unlocks In-Language Reasoning
Cohere discusses multilingual reasoning: when a model explains a wrong answer in a language you do not speak, data mixing is how they unlock in-language reasoning.
Why it matters: Teams building multilingual models can read how data mixing keeps reasoning in the user's language.
Monday, October 5, 2026
Feature North 2: Enterprise AI without compromises
North 2 is Cohere's biggest upgrade yet, combining enterprise-grade security, intelligence, cost governance, and full-stack control in one agentic AI platform that runs wherever your data lives.
Why it matters: Enterprises can run an agentic platform with security and cost controls without moving their data.
Feature Cohere and PwC partner to accelerate secure enterprise AI adoption
Cohere and PwC are partnering to accelerate secure enterprise AI adoption.
Why it matters: Companies working with PwC can adopt Cohere under a partnership aimed at secure enterprise use.
Wednesday, September 30, 2026
Feature RCP-nDCG@10: A more complete way to measure retrieval relevance
Cohere introduced RCP-nDCG@10 as a more complete way to measure retrieval relevance.
Why it matters: Search teams can score retrieval with RCP-nDCG@10 when a single relevance number is not enough.
Model Important Introducing Embed 5—A new family of frontier embedding models
Cohere introduced Embed 5, a new family of frontier embedding models.
Why it matters: Search and retrieval builders can evaluate Cohere's new Embed 5 embedding models.
Friday, September 25, 2026
Feature Compass is coming to the cloud
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.
Tuesday, September 22, 2026
Feature AI change management: A human-centric approach
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.
Wednesday, September 16, 2026
Feature Cohere and OpenText partner to bring trusted agentic AI to governments and regulated industries
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.
Feature Cohere and Aleph Alpha sign agreement to become the first transatlantic sovereign AI solution
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.
Sunday, September 13, 2026
Feature Who Gets to Define the Rules for AI?
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.
Thursday, September 10, 2026
Model Important Introducing North Small Translate: A leading sovereign open-weight machine translation model
Outsized performance, right-sized footprint — translation built for speed and cost efficiency.
Why it matters: A new model option to evaluate — compare quality, price, and context length against what you use today.
Tuesday, September 8, 2026
Model tiny-aya-base-32K
Cohere published the tiny-aya-base-32K model on Hugging Face for text generation (LLM). License: cc-by-nc-4.0.
Why it matters: Weights are downloadable for self-hosting or fine-tuning — read the model card and license before production use.
Feature Inside the megakernel serving engine for North Mini Code
Official feature update from Cohere (Cohere Blog).
Why it matters: A new product capability you may be able to use right away, without code changes.