Hacker News new | ask | show | jobs
by thunderbird120 166 days ago
That's what it's running on. It's optimized for very high throughput using Cerebras' hardware which is uniquely capable of running LLMs at very, very high speeds.