Dominik Filkus (@DominikFilkus) header
Dominik Filkus (@DominikFilkus) profile picture

⚙️ Tech lover | 🖥️ Software Engineer | 🕹️ Retro games to AI innovation

Joined November 2015
Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)Photo by Dominik Filkus (@DominikFilkus)
Opus 5.5 is quite good at creating 3D games with Three.js! The controls are smooth, the graphics are eye-catching, and the game itself is enjoyable. This is the best version of this game I have generated with a model using only one prompt.
3
7
1,385
The Microsoft outage is unbelievable. Outlook and other products have been down / unstable for more than 12 hours.
546
T3 Code is like an exciting open world RPG. You just go exploring, mess around a bit and find useful, sometimes rare items on the way. When I think I explored all the features of it, I still find more, not to mention frequent updates. It's rare to find such a good software nowadays as @t3dotcodes
111
It's funny that we can see this kind of estimate from SOTA models nowadays.
1
113
Dominik Filkus retweeted
I did say we'd be releasing our models periodically...introducing Gemini 4 Carbon, our next frontier model. Following Argon's release just two weeks ago, Carbon shows the pace at which we’re pushing the frontier, it ranks #1 on the recently updated v5.0 of @ArtificialAnlys Intelligence Index at 59. It also leads across key cybersecurity evaluations scoring 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching. We’re seeing significant improvements in long-horizon tasks even compared to Argon, and we’ve strengthened frontier safeguards. As we did with Argon, we’re going to roll it out gradually + responsibly. Some trusted cyber defenders will get access starting this week. We'll be expanding it to more partners through our Fairwind Program, ahead of wider availability soon, so hang tight!
2,068
2,732
1,920
33,985
7,001,713
Dominik Filkus retweeted
We’re rolling out Gemini 4 Argon in Antigravity to paid Google AI Pro subscribers today. Demand has been incredible and we've scaled capacity deliberately to get here. We’ve rolled out a separate weekly-only Argon quota for Pro on top of the existing Gemini bucket for 3.8 Flash so you have the best of Gemini for all your tasks.
1,070
1,323
1,027
15,149
1,512,701
Dominik Filkus retweeted
Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.
1,980
4,609
4,525
42,685
9,062,217
Dominik Filkus retweeted
DevDay livestream in 2 hours. We built some great stuff for you!
898
374
225
9,448
784,398
Dominik Filkus retweeted
Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
1,939
4,434
3,032
55,596
12,811,653
One of the best features I love about Opus 5.5 compared to Fable 5.1 is how the model answers. I am used to writing skills / prompts in a way that I would expect more human / less sloppy answers. Now Opus 5.5 does this natively, it communicates more like a human. It does not mention deep technical details where those are not important and answers shortly, getting to the point more quickly.
1
372
It might be too early to judge but today I had to switch back to Fable 5.1 because I felt like Opus 5.5 just overcomplicated things and failed on easy tasks without explicit coordination.
211
One of the reasons I liked Grok 4.6 (besides Grok Build) was the price. Grok 4.7 seems more expensive than GPT-6 Astra and slower than the previous version. I can't really see why anyone would use it over 4.6, the intelligence gap is just not that huge.
163
But they wanted to slow down... 😂
Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.
2
263
Dominik Filkus retweeted
Replying to @ThePrimeagen
I am already seeing the web falling apart slowly. Small alignment issues here, non functional buttons there, errors happen more frequently. And I am not talking about underground software but software that everyone uses.
1
3
430
Dominik Filkus retweeted
74
977
57
9,883
350,481
Dominik Filkus retweeted
Mark Zuckerberg: “Meta delayed shipping Muse for several months to focus on safety and security. We didn't call for everyone else to do this before we would. We just did it as part of our day-to-day work because it was clearly the right thing for people and for us.”
Last month I wrote about how we can build a positive and safe future for everyone: meta.com/thefutureisforevery… Every lab has the responsibility and incentive to move at the pace required to train its models safely, and the ability to take its own actions to ensure that happens. The reality is: - People won't want to use agents that are misaligned with them and that don't do what they ask, so labs have a strong natural incentive to make their models more aligned. There is a lot of debate about slowing progress on capabilities until alignment catches up. My view is that trust and alignment are quickly becoming the most important capabilities that will differentiate agents and models. Any lab that doesn't focus on alignment will fall behind. - Labs face significant liability if their models cause harm, so they have a strong incentive to prevent this as well. Meta delayed shipping Muse for several months to focus on safety and security. We didn't call for everyone else to do this before we would. We just did it as part of our day-to-day work because it was clearly the right thing for people and for us. I'm proud of the security foundations we've built. - Engaging independent evaluators and advisors is industry best practice. MSL already does this today in several areas because it helps produce better work. Other labs can just do this too. In general, it would be helpful for there to be a larger and more diverse ecosystem of evaluators. - Committing the significant majority of compute towards serving people rather than racing towards recursive self-improvement is one of the best ways to ensure we develop this technology safely. Meta has made this commitment and other labs can do this as well. I believe the key to building a positive future for everyone is maintaining the right balance of power. This is within our power to do.
384
1,153
80
17,066
1,187,813
GPT Image 2.5 still has no clue what a C64 looks like :/
3
314
Dominik Filkus retweeted
This is GPT-6 Astra. Anything you can do on a computer, Astra can do for you. Fast.
9,138
35,066
26,631
337,717
139,540,501
"Fable 5.1 will cost an estimated 25% less than Fable 5 for typical workloads... For highly agentic work, the savings will often be much larger—up to approximately 45%." 🙃
1
3
276
Dominik Filkus retweeted
We’re introducing Claude Fable 5.1 and Claude Mythos 5.1. They're the world’s most advanced models for coding and knowledge work.
2,850
7,255
4,816
65,176
21,151,310