frost

SWE

[frost @fr0stGFX]

🫡

[Tibo @thsottiaux]

Tomorrow we will bring back the 5h limit for Plus accounts across ChatGPT Work and Codex. I had mentioned this a while ago, but then postponed it.

This is necessary as (a) the 5h limit allows us to smoothen the load on our compute, allowing to keep the plan generous in terms of weekly usage and (b) users on the Plus plan are relatively casual and new users, but then also just accidentally eat through their whole weeks usage and then are confused, making it not a great experience.

We are for the upcoming months keeping the 5h limit not enabled for Pro $100 and Pro $200 subscriptions.

[frost @fr0stGFX]

OpenAI is previewing Private Safety Processing.

Better safety for autonomous AI without giving OpenAI personnel access to the underlying content.

Privacy and safety both matter more as agents get more capable.

[OpenAI @OpenAI]

We will continue to offer Zero Data Retention for frontier models.

As AI takes on longer, more autonomous work and delivers greater value to businesses, safety systems also need to identify risks across related interactions.

To help address those risks, we're previewing Private Safety Processing, which is designed to improve safety without giving OpenAI personnel access to the underlying content.

[frost @fr0stGFX]

Heads up if you use Daybreak Blue

A technical issue removed access for a limited number of users, and those affected will need to re-verify.

Check your email if your access disappeared.

Also, Advanced Account Security becomes required for everyone on September 1.

[OpenAI Developers @OpenAIDevs]

Due to a technical issue, a limited set of users’ access to Daybreak Blue is no longer active and they will need to re-verify to maintain their access.

This isn’t the experience we want to deliver, and we’re making sure that the issue doesn’t repeat in the future.

Affected users have been notified by email with a path for verification and our teams are on standby to help. If your access is uninterrupted, you don’t need to do anything right now.

As a reminder, all users will be required to have Advanced Account Security by September 1 to maintain Daybreak access.

[frost @fr0stGFX]

I genuinely would like to know if Grok 4.6 backs up the hype.

Tell me why I should switch from codex

[frost @fr0stGFX]

OpenAI paused Reinforced Learning training on its latest models for two weeks.

AI is moving fast. Kinda scary when the people building it decide it’s time to hit pause.

[OpenAI @OpenAI]

As models become more capable, the risks associated with developing and testing them internally also grow.

We temporarily paused reinforcement learning (RL) training on our latest models intended for deployment for two weeks while we hardened and red-teamed our research environments and expanded monitoring coverage.

Our largest planned frontier RL run remains on hold while smaller-scale training and evaluations validate these safeguards and establish more evidence of alignment.
https://t.co/ecbMMmVoox

[frost @fr0stGFX]

Why is everyone cancelling their Claude subscriptions suddenly?

[frost @fr0stGFX]

@lunardreamn $200 codex subscription

[frost @fr0stGFX]

Don’t look at the reply’s lol

[Bojan Jakimovski @Shekswess]

Usage limits have been reset for all paid ChatGPT Work and Codex users.

Happy Monday you all. Hope it is a fantastic week.

[frost @fr0stGFX]

GitHub’s API requests, GIT operations, webhooks, and more are seeing degraded performance

[frost @fr0stGFX]

Anthropic has been surprisingly quiet.

Are they dropping something new soon? 🤔

[frost @fr0stGFX]

Pretty cool. You can enable a 1M-token context window in Codex with GPT-5.6 Sol.

Could be really useful for larger codebases and longer sessions where you want Codex to keep more context around.

Nice option to have

[Tibo @thsottiaux]

Here is how to enable a 1M-token context window in Codex for GPT-5.6 Sol.

Even though we have tuned the context limit in Codex to be set optimally when it comes to performance and cost, this is a common ask, so here it is documented.

A larger context window lets Codex retain more code, tool output, and conversation history before summarizing older material. You need a model that supports it. And GPT-5.6 Sol, for example, has a documented 1,050,000-token window.

Open ~/.codex/config.toml and add or update these settings at the top level, before any [section] headers:

```
model = "gpt-5.6-sol"
model_context_window = 1000000
model_auto_compact_token_limit = 900000
```

The first setting selects the model. The second tells Codex to use a one-million-token context budget. The third starts automatic history compaction around 900,000 tokens, leaving some headroom. Restart Codex client and start a new session after saving.

To try the configuration for a single CLI session without changing your defaults:

```
codex -m gpt-5.6-sol \
-c model_context_window=1000000 \
-c model_auto_compact_token_limit=900000
```

Have fun, but also know that we tuned the default carefully!

[frost @fr0stGFX]

@thsottiaux A reset would be great on a Sunday 👀

[frost @fr0stGFX]

Potential codex reset 👀

[Tibo @thsottiaux]

@Ananth7e Maybe

[frost @fr0stGFX]

What’s actually better?

Codex app or Codex CLI

Let me know why 👇

[frost @fr0stGFX]

@Lyracutevibes $200 AI subscription

[frost @fr0stGFX]

@Ananth7e Even worse when you realize you used your weekly limit

[frost @fr0stGFX]

@rezoundous Perfect time to launch 10 sub agents

[frost @fr0stGFX]

How is every 16 year old making $10k+/month with SaaS?

[frost @fr0stGFX]

2024: VS Code vs JetBrains.

2026: Codex vs Claude Code

[frost @fr0stGFX]

Thinking of paying for a coding AI subscription.

If you could only pick one right now, which would you choose?

- Cursor
- Claude Code
- Codex
- Copilot
- Antigravity