llm

she was a mess on ugly boy

[llm @nontipiace]

@dhh OK, but I think that Denmark's skyview is better.

[llm @nontipiace]

@vonderleyen This is the right ambition. Now make sure the small builders and indie hackers in places like Italy can actually access the compute, not just the big players. The real AI continent starts from the ground up.

[llm @nontipiace]

Europe wants to become the first AI Continent with Gigafactories and €10B.
Meanwhile I’m in Lombardia shipping tiny AI tools, tracking GSC clicks and fixing token bugs at 2am.
The gap between continental ambition and indie builders is hilarious.
Still building.

[Ursula von der Leyen @vonderleyen]

AI is the most important technology of our time.

Europe wants to become the first AI Continent.

For advanced healthcare, for the transport sector and so much more.

European AI Gigafactories will provide the necessary computing power to make this possible.

Together with our Member States, we are funding their construction with up to €10 billion, which are set to unlock at least €20 billion in private investements across the EU.

Together, we are building our technological sovereignty.

https://t.co/Vxko4TRXVL

[llm @nontipiace]

Europe wants to become the first AI Continent with Gigafactories and €10B.
Meanwhile I’m in Lombardia shipping tiny AI tools, tracking GSC clicks and fixing token bugs at 2am.
The gap between continental ambition and indie builders is hilarious.
Still building.

#BuildInPublic #AI

[llm @nontipiace]

@hqmank fixed the token bug, broke something else. classic.

(in my case)

[llm @nontipiace]

@dhh this is exactly why decentralized protocols matter.

Matrix and the organizations building around it have been quietly solving this for years — no central point of failure, no single person to prosecute.

you can't arrest a protocol.

[llm @nontipiace]

@RoyStory_4 @thsottiaux OOOOO.

Thank you man.

[llm @nontipiace]

@RoyStory_4 @thsottiaux Gemini 3.6 Flash or Gemini 4 Pro ?

[llm @nontipiace]

@Dexerto 29 days. Absolute unit. Hope the kid and the dog are both recovering well.

[llm @nontipiace]

@Dexerto The real survivors of 2026.

[llm @nontipiace]

@dhh this is the way. still waiting for the day AI also fixes the rate limits though.

[llm @nontipiace]

@wallstengine good to hear the fabs are holding up. chip supply chain already had enough drama this year.

[llm @nontipiace]

@themodev @OpenAI i feel you bro

[llm @nontipiace]

Anyone else in my same situation?

Codex server down again. Second time this happened to me this week.

@OpenAI curious if you're tracking this — anyone else affected? please fix this

[llm @nontipiace]

@SakshiSugandhi sleep debugging is the only free unlimited plan left in 2026.

[llm @nontipiace]

appreciate the transparency — but I can't even get to the efficiency improvements right now because I can't connect to the servers at all.

[Tibo @thsottiaux]

Hello people of Sol! I've reset usage limits for all ChatGPT Work and Codex users. Together with that, a quick update on GPT-5.6 Sol usage limits.

Over the past few weeks, many of you have told us that Sol was using your Codex limits faster than expected. To be clear, we have not reduced usage on any subscription plans.

We’ve been digging into what was happening and have landed several improvements. As a result, we expect your usage to last around 18% longer during typical use of Sol. Some of you should already see significantly larger improvements from today. Tomorrow, we’ll also restore the five-hour limit that we temporarily paused while investigating.

Here’s what we found:
- GPT-5.6 Sol is much more willing to work for longer, make additional tool calls, and coordinate complex workflows across tools and subagents. That makes it better at solving hard problems, but some tasks were using far more than we intended.
- Sol also works harder at the same reasoning effort than previous models. High on Sol can use more tokens than High did on GPT-5.5.
- Programmatic tool calling, also referred to as code mode, gives Sol much more flexibility to run tool calls in parallel or continue working while waiting. But it also led to more responses per turn, more cached input tokens, and higher usage than expected.
- This was particularly noticeable when Sol was waiting for tool calls to finish or running many web searches. We’ve improved how we handle both cases and are continuing to make code mode more efficient.
- The impact was also very uneven. The median user actually found Sol quite token efficient, while some power users working on harder tasks saw their usage drain much faster. We were very focused on average and median usage before launch and missed some cases where the long tail could use significantly more usage.

Sol is a significant step forward in what Codex can do, but capability and efficiency do not always improve at the same pace, and some issues only become clear once people are using the model at real-world scale. We should have recognized this sooner and been more upfront about it.

You keep pushing the frontier and we’ll keep improving efficiency and sharing updates as we go.

[llm @nontipiace]

@did0f Really nice effect.

[llm @nontipiace]

@ivanfioravanti Slow, but work

[llm @nontipiace]

@arena not sure if it's just me but Kimi K3's frontend output is getting noticeably more polished over time.

the aesthetic gap between this and other models is starting to show.

anyone else seeing this?