kopiro

[Theo - t3.gg @theo]

Maria is still under 6k followers, good opportunity to invest early

[antirez @antirez]

It is my belief that many devs right now are not maximizing what they can do with automatic programming because they still look at the code. Doing it makes you the bottleneck. Your time is better invested in new ideas, QA, design, and asking yourself what is your goal.

[Mario Zechner @badlogicgames]

looks like we've entered the phase of depression in our trade.

those who've gotten their ai psychosis out of their way in 2025 welcome you. cookies are over there. coffee is free too.

[kopiro @kopiro91]

@fregante Pretty common in video games :)

[kopiro @kopiro91]

I made SoundAnchor for a simple Mac annoyance: your AirPods sound great, then you join a call and suddenly everything sounds terrible.
It keeps your Mac using the right microphone, and can also play the same audio through two headphones or speakers.
https://t.co/8Fjfy2rj2j

[Vito Botta @vitobotta]

September 15. That's the deadline Cloudflare gave AI companies. Separate your crawlers for search vs training vs agents, or get blocked by default on ad-supported pages. Three categories, three separate policies. Free plan included. BotBase gives Enterprise customers a searchable bot database too.

https://t.co/G3gHtWDnJm

[kopiro @kopiro91]

@jxnlco Tokens usage

[kopiro @kopiro91]

@evilsocket What would you do?

[antirez @antirez]

Opus at this point is at a class similar to Kimi and GLM indeed. Many people switched. The real SOTA models are Fable, GPT 5.5, and the upcoming GPT 5.6. Those models are what make a difference between open models and closed ones. Opus can be marginally better than open models, and that is it.

[Tibo @thsottiaux]

Reminder that you can use the Codex App, CLI and SDK with any open source model, not just with OpenAI models.

https://t.co/spPifB4ck3

[antirez @antirez]

People often have this idea of Chinese models being N months behind US models. This mental model is not helpful to predict the future. The lag is due to compute deficit, so the playfield is that. It's not by chance that OpenAI and Anthropic have aligned models for two years.

[antirez @antirez]

Opus 4.8 is a total disaster. The problem is not the model per-se, they have Mythos and can anyway train a better model. The problem is: what is it happening inside Anthropic, at the management level? Since this is a product failure. If there was a technological issue NOT delivering is better.

[Mario Zechner @badlogicgames]

very accurate!

[Vito Botta @vitobotta]

OK, I finally got sold on ChatGPT Pro.

I kept reading people saying GPT-5.5 is better than Claude for some things, and that the $200/mo plan has limits so high that weaker open models stop making sense if you use agents a lot.

So I tested the $20 plan first. OMG. A few runs with Hermes Agent and OpenCode, and I upgraded right away.

The other part is Anthropic. I hate that Claude subscriptions are basically locked to Anthropic's own tools. No OpenCode, no Hermes Agent, no custom harnesses.

Security work is worse. You can apply to be whitelisted, but it doesn't seem easy, and I know people who got whitelisted and banned again soon after.

OpenAI seems less formal there too, from what I've read, but at least I can actually use the subscription in my workflow and my tools of choice.

[antirez @antirez]

The Internet just discovered GPT is better than Opus 🤨 Would love to understand the process that allowed the obvious to take months, otherwise my trust on the average AI expertise, which is already very low, may drop further.

[Peter Steinberger 🦞 @steipete]

autoreview is the most impactful skill I've added to my stack (next to https://t.co/SEj2XRpaD1). It automatically reviews your code before landing a PR.

Finds so many edge cases.
Sometimes it runs for hours.

https://t.co/zbUjIS2e1i

[DHH @dhh]

I've had more "I can't believe it's this good" moments with GPT5.5 than any other model since Opus 4.5. It's shockingly, scarily capable. Days and days of amazing progress. All steering, no handwriting. Yet utterly delightful to conduct its coding. So, so good.

[Sam Pullara @sampullara]

this is the most ridiculous prompt injection I have seen in the wild

[HealthFit @HealthToFit]

New update 12.9 is now available on the App Store.

It introduces the all-new Split Variation Chart — a fast, glanceable way to visualize your automatic splits.

No more boring multi-column tables. Your pacing tells a story instantly.
#LessNoiseMoreInsight