Loading activity…
@cerebras
The world's fastest AI inference and training. Try the latest open models at: https://inference.cerebras.ai/
AI
Cerebras says faster inference gives a harness room to inspect results and recover sooner while you steer the work.
AI
Cerebras says its assistant used Qwen 3.8 27B running at about 1,500 tokens per second. It compared the assistant with Grok Bot, Meta Muse and Claude Cowork on the same dinner-reservation task.
AI
Cerebras says CS4 uses the same wafer as CS3. A behind-the-scenes discussion with Angela Young, its SVP of product, explores where the extra speed comes from.
AI
Cerebras says Money Agent can turn a home-buying question into a real-time conversation, then into a financial goal.
AI
Connecting chip dies and working around defects were fundamental challenges, the company says, with power, cooling and reliability also crucial.
AI
Fleet of six open models spans 0.9 billion to 375 billion parameters.
AI
Sarah Chieng discusses Devin's reliability gains and enterprise use in a Cerebras video.
AI
Cerebras announced CS-4 with three wafer-scale engines and claims of faster inference.
AI
The mode launches first via the OpenAI API to a select group of customers.
