GPT-5-Codex
A proprietary GPT-5 variant optimized for agentic software engineering, long-running code changes, testing, review, and tool use across Codex surfaces.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why this model scores 79.8
GPT-5-level reasoning and specialized coding training support broad software tasks, with reported successful runs beyond seven hours. Availability across cloud, terminal, IDE, GitHub, mobile, and API gives high reach. High biological and chemical capability treatment and stronger cyber performance raise misuse, while default sandboxing, disabled networking, monitoring, prompt-injection training, and human review reduce residual control difficulty.
News tied to GPT-5-Codex
The model score of 79.8 rates this model's risk profile. The overall Doom Index of 67.9 measures the complete temporally weighted evidence record. These values answer different questions.
Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.
GPT-5-Codex extends agentic coding runs beyond seven hours
OpenAI released GPT-5-Codex across Codex cloud, terminal, IDE, GitHub, and mobile surfaces. The company reported successful independent software-engineering runs lasting more than seven hours, while its system-card addendum documented default sandboxing, disabled network access, prompt-injection training, and High biological and chemical capability treatment.
- Full item contribution
- +0.29
- GPT-5-Codex equal share
- +0.29
Model score history
-
R1
Doom Score 79.8
New exact public model verified from its dated launch page and system-card addendum.
14 Aug 2026