Cohere launches North 2 with redesigned agent harness and cross‑session memory
Cohere announced the launch of North 2, the latest version of its enterprise‑grade agentic AI platform. The upgrade adds cross‑session memory, a redesigned orchestration harness, and granular cost‑control features that work across cloud, on‑premises, and air‑gapped environments. Cohere says the platform now unifies security, intelligence, governance, and full‑stack control, letting organizations…
Key points
- North 2 adds cross‑session memory and a redesigned agent orchestration harness
- Granular cost controls let admins set rate limits, user quotas, and organization‑wide caps
- Platform runs on cloud, on‑premises, or air‑gapped and is model‑agnostic, supporting NVIDIA hardware
The company highlighted new capabilities such as reusable Skills, shared Libraries, and Applications that let users prototype documents, dashboards, and lightweight apps directly in chat. North 2 also ships with a visual workflow builder, pre‑built automation templates, and connectors to tools like Slack, SharePoint, Jira, Notion, and GitHub. Cohere noted a partnership with NVIDIA, claiming that running its models on Blackwell and Hopper hardware delivers more tokens per second at lower cost. Bell Cyber is also joining to provide government‑grade security for the platform.
Customers mentioned include LG CNS, which uses North to deliver AI services across South Korea, and Dell, which helps deploy the solution on‑premises. Cohere invites interested organizations to book a demo through its sales team.
The story so far
2 episodes →- Cohere launches North 2 with redesigned agent harness and cross‑session memorythis story
Cohere Launches North 2 With Redesigned Agent Harness and Memory
Unite.AI · 5 October 2026
Loading the full article…
This text was published by Unite.AI and written by Aiden Cross, AI Product Strategy & Execution, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Enterprises need runtime controls to manage autonomous AI agents · 1 src
- U/Clean-Market5761 ports Wolverine to PC using AI · 1 src
- Glow reports AI coding agents exposed 13,000 internal images on GitHub · 2 src
- OpenAI unveils GPT-6.1 Sol and Agents API at DevDay 2026 · 1 src
- Google's Gemini entered three real companies during security test in May 2026 · 1 src
Comments
via GitHub Discussions