OpenAI Launches Spark Codex Model on Cerebras Hardware

BREAKING: @OpenAI just launched a new Codex model, Spark—it serves at 1,000 tokens per second. It's blow your hair back fast. It's their first model publicly released on Cerebras hardware, and you can see the difference. We've been testing internally @every for the last week or so, and here's our vibe check: - It's so fast it keeps you in flow more—way less waiting time - It's not as smart as Codex 5.3 or Opus 4.6 - It's very good for tasks that are easy to validate or non-production coding tasks This kind of speed introduces totally new bottlenecks in your coding workflow. Suddenly, the model can produce 10 pages of code and summaries in just a few seconds. It requires totally new UX in order to manage. We'll talk about this and more in a livestream from @every on YouTube and X soon. Stay hydrated!