Latest from Cerebras: AWS, Cafe Compute, and Best Place to Work

Discover fast AI with Cerebras and AWS ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏  ͏ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­  

Amazon is deploying Cerebras CS-3s in AWS data centers   The world’s fastest inference is coming to the world’s leading cloud. Available via AWS Bedrock, the new service will offer leading open-source LLMs and Amazon’s Nova models running at the industry’s highest inference speed.    In addition, AWS and Cerebras are collaborating on a new disaggregated architecture that pairs AWS Trainium with Cerebras WSE to deliver 5x more high-speed token capacity in the same hardware footprint. Read More Cerebras Café Compute was a standing-room-only affair at GTC this year. Packed rooms. Lines around the block. From San Jose to HumanX to London, Café Compute has become a global hub for developers.   We explored a key shift in AI infrastructure: GPUs aren’t the only path forward—Cerebras is redefining what fast looks like.   Why? Because everyone wants to go faster.   Join us for future Cafe Computes all around the world!  RSVP Cognition Launches SWE-1.6 in Windsurf, Powered by Cerebras Inference   Cognition has released SWE-1.6, its latest software engineering model, now generally available in Windsurf. The model scores 50.4% on SWE-Bench Pro, beating GLM-5 and Kimi K2.5, while also improving UX by reducing common agent friction like overthinking, looping, and inefficient tool use. Cognition partnered with Cerebras to offer a premium tier running at 950 tokens per second for paying users, nearly 5x faster than the free tier, demonstrating how wafer-scale inference translates directly into better developer tooling experiences. Recognized by Fast Company and Forbes   Cerebras has been named to Fast Company’s Most Innovative Companies 2026 and Forbes America’s Best Startup Employers 2026. The rankings reflect both the technology we’re building and the team behind it. We're Hiring Thank you for being a part of our growing community! 👉 Get your API key 👉 Book time with the Cerebras team 👉 Follow us for updates Cerebras Systems, 1237 E Arques Ave, Sunnyvale, CA 94085 Unsubscribe Manage preferences Cerebras Systems, 1237 E Arques Ave, Sunnyvale, CA 94085 Unsubscribe Manage preferences