Cohere

Our secure agentic AI platform, North, is now widely available.

Say hello to the Cohere Centre! 🇨🇦 We’re proud to partner with Ottawa’s premier convention and event facility as it enters an exciting new chapter. The Cohere Centre will serve as a hub where leaders from business, government, technology, and the community come together right in the heart of Canada’s capital.

🚀 Sovereign AI for the world. Cohere & Aleph Alpha form transatlantic AI powerhouse anchored in Canada & Germany! Combining our global scale with European R&D excellence to build sovereign, enterprise-grade AI. Security, privacy & trust for businesses & governments worldwide. #SovereignAI #AIPartnership Learn more: businesswire.com/news/home/2026… Image from left to right: Rolf Schumann, Schwarz Digits, Samuel Weinbach, Aleph Alpha, Aidan Gomez, Cohere, Minister Solomon, Canada, Minister Wildberger, Germany

Excellent research by Conway Zhu, Ali Edalati, and Zewen Shen. Read the blog here: cohere.com/blog/vllm-inte…

Excited to share our work on production-ready W4A8 inference, now integrated in vLLM! By combining 4-bit weights (low memory) with 8-bit activations (high compute), we hit the sweet spot for both decoding and prefill — up to 58% faster TTFT and 45% faster TPOT vs W4A16 on Hopper.

Enjoyed the read? If you have deep experience in ML frameworks (training or inference) and love working on problems like these, our team is hiring! ML Systems Engineer, Frameworks & Tooling: jobs.ashbyhq.com/cohere/c99e61c… Audio Inference Engineer, Model Efficiency: jobs.ashbyhq.com/cohere/e912d84… Member of Technical Staff, Model Efficiency: jobs.ashbyhq.com/cohere/2a98903… Staff Research Engineer, Model Efficiency: jobs.ashbyhq.com/cohere/c80f0fe…

For real agentic workloads (North), short-context calibration wasn't enough. We calibrated AWQ on long internal agentic traces (up to 64k tokens) and added token masking in llm-compressor to exclude repetitive chat templates/tool descriptions from calibration stats. Plus QAD (quant-aware distillation) to close the last gap — matching the quality of our BF16 MoE model with W4A8.

Get more from speculative decoding in MoE models cohere.link/Et2rbsB

New Technical Report from @EkagraRanjan: Contrary to what you might expect, MoE-based LLMs make speculative decoding even more effective. Read more on our blog:

Was super interesting to chat with one of @cohere's senior PMs about its AI speech-to-text model, which it hopes to incorporate into its North platform soon.

"We're not looking for the easiest path, we're looking for the right path." NEW: Aidan Gomez (@aidangomez) is trying to build a very different kind of AI company at Cohere. Based in Toronto, $6.8B-valued @cohere works with G7 gov'ts and corporations to build everything from email summaries to ship telemetry apps -- while staying out of the consumer fray. "Privacy and security is essential, independence is essential, and the market has moved towards us," he says. On The Upstarts Podcast, Gomez talks about his journey from small-town Ontario to co-writing the Transformer paper that underpins Gen AI; why he sees AI as the best shot at restoring global growth and democratic resilience; and why it's important for him to take Cohere public in an IPO. Plus, he shares his Upstart Moment: building on ‘hard mode’ by not taking money from Amazon, Google or Microsoft early on. Catch the full episode, presented by our partner @mercury 🫡 youtu.be/RZ9I_CsDaoo

We're proud to be on the @Forbes AI 50 list again! It reflects our focus on building secure, sovereign AI that helps enterprises and governments put their data to work on their terms. Learn more: forbes.com/lists/ai50/

The future is open. At GTC Live, @cohere Co-founder & CEO, Aidan Gomez describes how open models give developers the autonomy they need to scale 📈, customize and innovate their way into the future of agentic AI. Watch the full #NVIDIAGTC keynote pregame: nvda.ws/4tdUFql

We worked with @RWSGroup to fine-tune our Command translation model, which improved language and cultural expertise. Now, that model is the “brain” that powers RWS’ Language Weaver AI translation solution.

Register now 👉 cohere.link/CBqUUv0

How can enterprises accelerate decision-making and improve efficiency? We're joining @Microsoft's Model Mondays on April 6th to discuss how AI can help you: 🌎 Drive real-world enterprise value 💡 Surface precise, verifiable insights 📌 Use Microsoft Azure and Cohere enterprise models to accelerate your AI journey

Learn more: cohere.com/blog/ensemble-…

Cohere is excited to announce an expansion of our partnership with @EnsembleHP to develop the healthcare industry's first revenue cycle management (RCM)-native large language model. This fully custom model will be informed by RCM insights and purpose-built to manage the complex financial workflows that support healthcare operations. We look forward to continuing to apply Cohere technology to reduce administrative burden in industries where it matters most.

Cohere Transcribe is setting a new standard for automatic speech recognition model accuracy in real world conditions – even with a noisy blender running. Try it out for yourself 👇 cohere.link/hpqiVWT