Hey everyone,
I wanted to share some exciting updates that OpenAI announced today regarding Codex usage. These changes should help extend your available usage and improve performance across different plan tiers:
What’s New
GPT-5-Codex-Mini — A more compact and cost-efficient version of GPT-5-Codex
-
Offers roughly 4x more usage compared to GPT-5-Codex
-
Comes with a slight capability tradeoff due to the smaller model size
-
Currently available in the CLI and IDE extension when signed in with ChatGPT
-
API support is coming soon
50% Higher Rate Limits for ChatGPT Plus, Business, and Edu plans
- This increase comes from efficiency improvements in GPU utilization
Priority Processing for ChatGPT Pro and Enterprise accounts
- Designed to deliver maximum speed for these tier users
How to Use GPT-5-Codex-Mini
The new Mini model is ideal for:
- Easier or more straightforward coding tasks
- Extending your usage when approaching rate limits
There’s also a helpful feature where Codex will suggest switching to Mini when you reach 90% of your limits, helping you continue working without interruptions.
Try It Out
You can start using GPT-5-Codex-Mini right now with:
$ codex -m gpt-5-codex-mini
Share your experience with the community:
- Have you tried GPT-5-Codex-Mini yet? How does it compare to the full model for your use cases?
- For those on Plus/Business/Edu plans, are you noticing the 50% rate limit increase?
- Pro/Enterprise users, how’s the priority processing performing for you?