StannyBoiii

@Dimillian For daily work, I have to use the ChatGPT desktop app to record meetings. It smokes everything else as prompting afterwards to pull actionables is a game changer. Way rather it be on Codex as the gpt app sucks compared.

What do you want next?

@ericmitchellai Yesterday it was being so lazy not engaging or proactive having to tell it constantly. Normally run terminal commands to get the job done instead it just gave me the commands to run

If you are experiencing any response quality issues with GPT-5.4 Pro, please let me know here! Have seen some chatter but no concrete issues

@thsottiaux Moving off it. Made the usage limits horrendous. Gone from occasionally hitting the 5hr limit to 4 prompts and it's gone. Moving to GitHub Copilot before they start reducing usage. You had something good🥲

Codex is for wizards

@LLMJunky It's all usage. I got a biz account and personal and no joke 3 prompts my 5hr limit window is gone even with 5.3 codex... My personal one I struggle to hit the 5hr window. Moving to GitHub Copilot before that gets limited.

🚨 OpenAI is moving Codex from message-based to token-based pricing for credits for all ChatGPT Plans in the coming weeks. As of right now, all Business & new Enterprise accounts have already started this new API token based system. The "legacy" system is a message based system, which is generally less granular. You get a fixed number of messages per credit (although some messgages can consume more than 1 credit). Token-based pricing changes that. You're billed based on actual input/output tokens consumed, so lightweight tasks cost less and heavy ones cost more. In other words, its API based usage rates. More predictable if you're watching usage, but also easier to burn through if you're not. Note: I am 98% sure this ONLY applies to credits, not your base Codex usage with your ChatGPT plan. If someone from OpenAI is reading this and could confirm this, would be great.

@brave W, fully switched back to brave cause of this.just need to sort out your vertical tabs

Brave now has an ultra dark theme! 🌑 To use it, first make sure you've updated Brave for desktop to v1.88. Then, go to Settings -> Appearance -> Theme and toggle "Ultra dark theme" at the bottom.

@sama You should win your users back by making a far superior product that helps the masses for good. ChatGPT has slipped, desktop app sucks, web app mid, codex is good but there's no middle ground for usefulness. Think Cowork but Codex and easy to share custom MCPs, skills, plugins.

I'd like to answer questions about our work with the DoW and our thinking over the past few days. Please AMA.

@embirico I use Codex for coding but I've also found it's super good to being an alternative to Claude cowork. Setup a couple skills, MCPs and automations that aren't code based and I have a way better platform to do my 9 to 5 than ever before, not a feature but maybe a direction.

Keep the feedback coming

@thsottiaux The IDE extension should have better GUI menus for setting config.toml stuff and MCPs etc like just buttons to enable/disable things hate having to look what I need to put in the config file

What do you hope to see Codex ship before end of month? Team is cooking with the power of a thousand agents and there is still time to redirect them.

@iannuttall W

Codex degradation report TLDR: - The investigation found no single root cause; rather a mix of behavior shifts and small bugs, with several fixes already shipped and more on the way. - Older hardware underperformed in evals and was removed, and improved load balancing is rolling out to reduce latency. - Compaction was used more often, and repeated compactions hurt accuracy; recursive summaries were blocked and a warning was added to keep chats focused. - The `apply_patch` flow could delete and then recreate files after failed diffs, which is risky if interrupted; mitigations are in progress and a longer‑term model fix is planned. - There was no fleet‑wide latency regression, but some sessions escalated retries and timeouts inefficiently; handling of long‑running and interactive tasks is being improved. - A constrained‑sampling bug produced out‑of‑distribution tokens, including occasional mid‑sentence language switches in under 0.25% of sessions; a fix is rolling out. - The Responses API added extra newlines around tool definitions; this had no measurable performance impact. - Evals show CLI 0.45 maintained quality while reducing tokens by roughly 10%; neither `--search` nor recent prompt changes caused regressions. - Infra analysis found low auth‑cache hit rates adding about 50 ms to tail latency; working‑directory error rates were unchanged. - The guidance is to prefer minimalist toolsets and shorter, targeted chats, and a dedicated real‑world performance team is being formed and hiring.

@thehealthb0t @grok is this true

Did you know this about sourdough bread?

@AravSrinivas @comet 100% ship that plus some minor customisation and it'll be my daily work and personal browser

Do you _really_ want vertical tabs on @comet?

@AravSrinivas @comet Only reason I won't use it as a daily driver is the lack of just core browser customisations even edge browser has more... Simple stuff like only show bookmark bar on a new tab. Vertical tabs, split screen and so on

What’s currently frustrating about Comet? @Comet will go through the comments here and triage issues for our team.

@AravSrinivas Brother 200 USD a month is insane if you want adoption offer it on the free tier get people hooked and they'll pay pro but no way 200 a month...

Introducing Perplexity Email Assistant: an agent that behaves as your personal/executive assistant on your email client (Gmail, Outlook); scheduling meetings, prioritizing emails, and drafting replies for you. Available to all Perplexity Max subscribers from today!

@perplexity_ai Amazing, but you need more than that

Perplexity Pro users can now connect their email, calendar, Notion, and GitHub to Perplexity. Enterprise Pro users can also connect Linear and Outlook.

@AravSrinivas Ship new android UI

Actually, fastest growing app on both App Store and Play Store

@AravSrinivas Ship new android UI already

Perplexity Finance is now available on Perplexity mobile apps: both iOS and Android. Just type in “finance” on Perplexity search bar and pick the Finance auto suggested option.

@AravSrinivas Android...

Another major Perplexity iOS app update. Team cooked. Answers are now streamed smooth as butter. Tables, markdown, intermediate steps. Update and enjoy!

@OpenAIDevs Huge W, was about to cancel and resub to Claude

We’re releasing new Codex features to make it a more effective coding collaborator: - A new IDE extension - Easily move tasks between the cloud and your local environment - Code reviews in GitHub - Revamped Codex CLI Powered by GPT-5 and available through your ChatGPT plan.

@ElonMuskAOC My feed is basically US politics and algo short form vids are the same and I’m not even in the US… gets boring quickly. Adjusted my interests and location based content that needs looking at

Do you wish the Twitter algorithm was better? If so, how?

@ElonMuskAOC Do it but leave out LinkedIn, instead of bots I’ll be slammed with random recruiters. Make sure it’s easy for the user! If I want long form YouTube I should be able to easily view my content. Same for all other types of ‘social’

Facebook. Instagram. Snapchat. YouTube. LinkedIn. WhatsApp. TikTok. Paypal. CNN. FOX. Amazon. All will become obsolete. X/Twitter will be the “all-in-one” app!