Key Points
- OpenAI launched Codex and the new GPT-5.3 Codex model for developers.
- GPT-5.3 Codex expands functionality from simple code writing to handling most developer tasks.
- The model can create complex games and apps from scratch in days.
- It runs 25 percent faster than the previous GPT-5.2 model.
- GPT-5.3 Codex was partially built using earlier versions of itself for debugging.
- Anthropic released its competing agentic coding model 15 minutes earlier than OpenAI.
- Both releases were originally scheduled for the same time at 10 a.m. PST.
Launch Overview
OpenAI introduced a new agentic coding tool named Codex, aimed at software developers, alongside an upgraded model called GPT-5.3 Codex. According to the company, GPT-5.3 Codex transforms Codex from an agent that merely “write[s] and review[s] code” into a system capable of performing “nearly anything developers and professionals do on a computer,” thereby broadening who can build software and how work gets done.
Performance and Capabilities
The firm reports that GPT-5.3 Codex can generate “highly functional complex games and apps from scratch over the course of days” after testing against a range of performance benchmarks. It also claims the new model is 25 percent faster than its predecessor, GPT-5.2. Notably, OpenAI describes GPT-5.3 Codex as the first model that “was instrumental in creating itself,” indicating that staff used early versions of the program to debug and evaluate its performance.
Competitive Context
The launch arrives shortly after Anthropic released its own agentic coding model. Both companies had originally planned to publish their tools at the same time—10 a.m. PST. However, Anthropic advanced its release by 15 minutes, slightly preceding OpenAI in the public rollout.
Implications for Developers
OpenAI suggests that the expanded capabilities of GPT-5.3 Codex could change the software development landscape by allowing a wider range of users to create sophisticated applications more quickly. The speed improvement and self‑debugging features are highlighted as key differentiators that may influence adoption among professionals seeking more efficient coding assistants.
Source: techcrunch.com