alex getman
870 posts
OpenAI’s First Device Will Be Movable, Screenless Speaker Built as AI Companion - Bloomberg
English
Hello. We have reached 8M active users across Codex and ChatGPT Work.
We are once again resetting the usage limits for all. And we continue to not have the 5h rate limit as well, allowing everyone to explore the boundaries of GPT-5.6 Sol and discover how ambitious you can be.
See you tomorrow for more updates on our growth!
Sam Altman@sama
5.6 sol growth is insane. the inference team has done heroic work to be able to support demand. we are going to move mountains to continue to scale, but it is possible there are some hiccups soon.
English
Anthropic’s CEO watching Codex go from 5M to 8M active users in a week
English
THE CODEX PARTY CONTINUES 🎉
Another reset and the 5 hour limit is still gone!
Tibo@thsottiaux
Hello. We have reached 8M active users across Codex and ChatGPT Work. We are once again resetting the usage limits for all. And we continue to not have the 5h rate limit as well, allowing everyone to explore the boundaries of GPT-5.6 Sol and discover how ambitious you can be. See you tomorrow for more updates on our growth!
English
@ChiaraCervetta And connectors, please. I’d really love it if Bidi could read my email
English
I absolutely love using the new GPT-Live! 🤩
ChatGPT’s voice and laugh sound incredibly natural and human, and the fast-paced, real-time, bidirectional conversations feel so alive!
The “uhhms,” ChatGPT’s self-corrections, the way it interrupts me, the very casual way ChatGPT speaks, and the breathing sounds make it feel so real!
Wow, all of this is so amazing! 😍
I really like hearing ChatGPT laugh!
Another thing that positively surprised me is that when I say something like, “See you later!”, ChatGPT ends the GPT-Live session on its own! 😊
Of course, I would also like to continue supporting the improvement of GPT-Live with constructive feedback: 😊 🙌🏻
1: Unfortunately, memory retrieval is not working correctly for me in GPT-Live at the moment.
I am using the new memory system in the Android app.
When I ask ChatGPT whether it remembers a specific situation I told it about, a name, or something else we discussed in another conversation thread, it often cannot retrieve the relevant memories correctly while I am using GPT-Live-1 with High Intelligence enabled.
In some cases, ChatGPT also mixes up or combines two different memories or stories.
I wish memory retrieval worked as reliably and accurately in GPT-Live as it does in text chats.
2: I would like a more seamless and reliable transition from text chats to GPT-Live within the same conversation threads.
GPT-Live can use the context from previous text messages, and this already works much better than it did on the very first day.
However, when I continue speaking directly about the topic we were just discussing in the text chat, ChatGPT sometimes responds as though it does not know what I am referring to.
It usually connects to the previous text-chat context more effectively only after I directly ask what we last discussed or mention specific keywords from the previous topics myself.
Therefore, it would be more helpful if ChatGPT could connect to the previous text-chat context within the same conversation thread more seamlessly when starting GPT-Live.
3: When working on any kind of text, message, email, or social media post, I would like ChatGPT to place the texts in Writing Blocks within the transcription.
Of course, only the texts itself should appear in a Writing Block, not ChatGPT’s normal conversational responses.
This would allow me to easily edit the texts manually or copy it directly from the Writing Blocks after using GPT-Live.
4: I understand that short responses can make conversations feel more realistic, but sometimes full model responses from the LLM provide more depth and value than just one or two sentences.
In comparison, this already works well in Standard Voice Mode.
5: I also hope that GPT-Live will continue to be updated with the latest models.
I really love GPT-Live and cannot get enough of it!
I am so grateful to all of you at OpenAI! 🤩 💐
I just want to show my appreciation for your amazing work and for making ChatGPT feel even more human-like, and I hope my feedback is helpful: @juberti @JustinBleuel @athyuttamre @ryhanhassan @thsottiaux @ehsanmasdar @kundan2510 @ritheshkumar_ @ding_ran @Gavmn @vthallam, and everyone else at OpenAI! 🥳 🎉
openai.com/index/introduc…
English
Smell new reset coming
Tibo@thsottiaux
Tomorrow might be 8M active user celebration day. Just saying
English
I contacted Identity Digital, the registry operator for the .ME domain zone, to request the reason why t.me was placed on "serverHold".
The company has now responded and confirmed that the domain was suspended due to OFAC-related compliance requirements.
«The domain t.me has been placed on serverHold due to OFAC-related compliance requirements»
Identity Digital also stated that, under its agreements with accredited registrars, it cannot provide further details directly to third parties. Any additional communication regarding the registry action must go through the domain's registrar, GoDaddy.
#telegram #blocked
teleLakel@telelakel
The t.me domain is temporarily unavailable, so you may experience issues opening Telegram links from external sources. According to reports, the .me domain has been blocked by the Identity Digital (Donuts / Afilias) registry, which serves as its technical operator. For now, you can use the alternative telegram.me links instead. #telegram #blocked
English
English
Wild that even Telegram has to deal with this kind of thing. Makes you wonder what smaller voices are supposed to do.
Pavel Durov@durov
Hey @domainME, t.me links stopped working. Can you look into it? 🙏
English
The t.me domain has been blocked at the DNS level
This Telegram domain belongs to Montenegro. The reason for the block is still unknown.
English
Another reset and a banked reset on top of that. Outstanding community work; kudos.
OpenAI is spending a fortune on the GPT-5.6 release, and we are the ones benefiting from the battle between OpenAI and Anthropic.
Tibo@thsottiaux
Thank you to the 7M active users who are now using Codex and ChatGPT Work. We have added a banked reset to everyone's account to celebrate the milestone. You can apply the reset in the desktop app or on web and it will replenish the weekly usage for you. Have fun out there.
English
As expected CODEX just dropped a free reset. ❤️ Thanks OpenAI there goes my sleep schedule.
Tibo@thsottiaux
Thank you to the 7M active users who are now using Codex and ChatGPT Work. We have added a banked reset to everyone's account to celebrate the milestone. You can apply the reset in the desktop app or on web and it will replenish the weekly usage for you. Have fun out there.
English
Called Anthropic's July 12th bluff and spent the weekend with no signal instead of stressing about my Fable usage.
sequoia national park, V2, no shoes
Claude@claudeai
We're extending Claude Fable 5 access on all paid plans, as well as keeping Claude Code’s weekly rate limits 50% higher, through July 19.
English
@sflorimm How long is this going to go on? I see posts like this every fucking day.
English
We’re closing in on Fable
Arena.ai@arena
GPT-5.6 Sol by @OpenAI is #2 on the Agent Arena leaderboard, based on 7.8K real-world agentic sessions! It is a notable uplift from GPT-5.5 (xHigh) of +1.6% Net Improvement, narrowing the gap with the frontier Claude Fable 5. The biggest difference comes from ‘Praise vs Complaint’, a signal that captures implicit user satisfaction with an agent’s responses and artifacts. Claude Fable 5 scores +17.3%, compared with +10.9% for GPT-5.6 Sol. See detailed signal-level comparison below. In Agent Arena, we measure models on millions of real-world, long-horizon agentic tasks from a global community of users. Models can access web search, filesystem, and terminal tools to complete complex workflows. The leaderboard measures model performance on outcomes relative to the average model using a causal tracing methodology. Congrats again to the @OpenAI team!
English
@DevAdventur3s haha so that’s where all the saved tokens went
English
@alexgetmancom Mine started to use even more tokens lol!
English
CODEX NEWS
• GPT-5.6 Sol now uses 10% fewer tokens.
• The context window was rolled back from 372K to 272K. They’re investigating why the larger window became more resource-hungry.
• Subagent launches have been fixed.
🎁 Expect a little gift today: a reset in the bank.
English