Codex is now available to all ChatGPT Plus users! Your most requested feature is here too! The agent can now access the internet while completing your tasks, and you control exactly which domains Codex can use.
@romainhuet
-
Structured Outputs Now Support Regex in Schema
By
–
Structured Outputs got a big upgrade, and you can now write regex directly in your schema!
-
Codex AI Agent Completes Legacy Migration in Minutes
By
–
Awesome use case for Codex: a complete legacy migration in one shot. What used to take weeks can now take minutes, handled entirely by an AI agent.
-
Quick Walkthrough Guide for Setting Up Codex Environment
By
–
Ready to dive into Codex? Here’s a quick walkthrough from @DKundel to help you set up your environment and kick off your first tasks!
-
Codex CLI Gets Better with New Codex-Mini Model for Code Editing
By
–
Codex CLI is also getting better today, with a new codex-mini model optimized for fast code editing! Check out the latest with @hwchung27 @fouadmatin @anupk24
: -
Codex-1 Achieves State-of-the-Art Performance on SWE-Bench Verified
By
–
The codex-1 model is state of the art on SWE-Bench Verified, we published the numbers on the blog. More importantly, we optimized it to generate code that people actually want to merge, not just code that scores well on benchmarks!
-
OpenAI Launches Codex: AI Agent for Parallel Software Engineering Tasks
By
–
Coding is about to change forever. We’re launching a research preview of Codex, a cloud-based software engineering agent that can work on many tasks in parallel. Codex is powered by codex-1, a version of OpenAI o3 optimized for software engineering.
-
Reinforcement Fine-Tuning: o4-mini Learns from Minimal Examples
By
–
Reinforcement fine-tuning still feels magical to me.
— Romain Huet (@romainhuet) 12 mai 2025
Compared to regular fine-tuning, RFT on o4-mini learns from just a handful of solid examples to improve model performance for your use case, even in complex domains. https://t.co/gqyrKW3DqiReinforcement fine-tuning still feels magical to me. Compared to regular fine-tuning, RFT on o4-mini learns from just a handful of solid examples to improve model performance for your use case, even in complex domains.
-
Romain offers assistance with o4-mini verification issues
By
–
Sorry you’re running into friction, Stefan! Can you please DM me your org ID (and any ticket #) and I’ll nudge the team to make sure you get the verification sorted. Can’t wait to see what you build with o4-mini!