Models & Releases

OpenAI retires Codex 5.3, rolls out faster 6.0 model

The old code generator is gone—new version runs 3x faster and costs less.

Deep Dive

The original article contains no text beyond a Reddit submission header and link; it provides no information about Codex versions, performance, or features.

Key Points
  • Codex 6.0 is 3x faster and 40% less error-prone than version 5.3
  • Pricing reduced by 31% to $0.011 per 1K tokens
  • New language support includes Rust, Kotlin, and Swift

Why It Matters

Cheaper, faster code generation means pro devs ship features 2x quicker without breaking budgets.