Emircan ERKUL

@svpino Thanks for enlightenment, followed

Google published an entire library of highly sophisticated, end-to-end agent examples. 100% open-source. • Complete documentation • Source code • Ability to one-click deploy In the video, I break down one of the coolest examples in this collection.

@holynski_ Okay we gave our respect incase you also want to represent yourself as a duck being not he or she or x, y, z but a duck 😂 - Muted!

My whole life, I've wanted to be an elephant riding a motorcycle through my hometown. Now, it's finally possible.

@YaBoyRoshi Sir, its part of the show :) Actually really great regarding presenting improvisation capability

IM CRYING! 😂😂

@earthcurated Bees are most probably do not know that, so i do not need to know that.

Today is World Bee Day.

@makingAGI Looking forward to see 17b 21b coding related hrm models. Also mlx support. It must scale if we call it a win.

The HRM-Text paper is now available 🎉 HRM-Text explores a different approach to language model pretraining: hierarchical recurrent computation, task-completion training, and latent-space reasoning. At just 1B parameters, HRM-Text achieves competitive performance with dramatically lower training cost and data requirements. 1B parameters 40B unique tokens ~1 day of pretraining ~$1000 training cost

@PhysInHistory Ah i see, its just gamification in your brain... Sure, keep chase the rat.

@PhysInHistory Lets say it has, what would you do? Lets say it hasn't, what would you do? You just can ask yourself what is the purpose of this question to not waste your time further. When answer become yes, or no and it do not change anything... why still waste your time?

Does the Universe have any edge? ✍️

@demishassabis @antigravity @GeminiApp Enjoy what? I tried and within 10m i consumed all credits as a pro subscriber

Gemini 3.5 Flash is amazing! - Performs better than 3.1 Pro on coding & agentic tasks - 4x faster than other frontier models - 12x faster in @antigravity - 800 tokens/sec! - Often at less than half the cost And Pro to come… Try it in @antigravity, @GeminiApp & more - enjoy!

@Google @antigravity Marked as spam! Why I have to see those? As you clearly mentioned, this feature is for only US.

Introducing Gemini Spark ✨ It’s your 24/7 personal AI agent that helps you navigate your digital life, taking action on your behalf, and under your direction. 🧠 It runs on Gemini 3.5 and is built on @Antigravity, so it can perform long-running tasks easily in the background. ⏱️ And because it runs on dedicated virtual machines on Google Cloud, you don’t even need to keep your laptop open. 🧰 Spark will integrate seamlessly with Google tools, and soon with third parties through MCP. #GoogleIO

@cerebras Could you provide cheap/free for consumers deepseek v4 flash one? at 1k speed?

Google's new fast model 3.5 Flash vs Cerebras

@cerebras @ArtificialAnlys @Kimi_Moonshot Great 🎊🎊🎊 Could you make Deepseek v4 flash faster? 5k maybe?

Cerebras is now running Kimi K2.6 – a trillion parameter model – in enterprise trials. At ~1,000 tokens/s, this is the fastest frontier model performance ever measured by Artificial Analysis @ArtificialAnlys.

@AhmedAlNeaimy @antigravity 😂

@antigravity THE MOST FUCKING ANNOYING POP-UP IN THE WORLD SINCE 6 MONTHS TILL NOW, IF I'M A GOOGLE CEO I'M GONNA FIRE THIS TEAM AND FIX IT MYSELF.

Introducing Antigravity 2.0, a new standalone desktop application that delivers fully on that original glimpse of a truly agent-optimized experience. Rebuilt from the ground up with multi-agent teams, scheduled tasks, native voice and one-click integration with other Google products. Learn how to get started with Antigravity 2.0 👇

@jcasconecta @antigravity Use with @zeddotdev

@antigravity What happened to the code editor? We had a great tool, and I genuinely preferred your platform over all the alternatives. Who is making these product decisions, and whose feedback are you listening to?... 😮‍💨 #SMH #ImNotMadImDisappointed

@OfficialLoganK @Google With 3.2Q Tokens we suppose to find cure to the cancer...

3.2 quadrillion tokens a month and still growing :)

@Google Looks good but its a bit expensive but really fast. But naming introduce confusion regarding price. Flash means fast not cheap. Maybe you could go with Gemini Fast (new naming) And Looking for Gemini Flash Fast one 😂

Meet Gemini 3.5 Flash — our strongest agentic and coding model yet. It delivers frontier-level performance at 4x the speed of comparable frontier models — often at less than half the cost. Generally available, starting today. 🧵 #GoogleIO

@antigravity Not a problem but could be a good extra... My AI coding workflow already settled with PI coding agent & @deepseek_ai models You just have to give cheap/almost free option to compete with deepseek pricing to collect data...

Antigravity's first Google I/O. Hold on to your keyboards.

@lexx_aura Imagine you bought gold yesterday at 4500 and today it become 2000... Yea it is same. You invest on something and hope to get benefit from it. IDK but if you're in middle of the uni, you still put your stop lost. Or wait till see the bottom or idk Note: not a professional advice

Imagine studying 4 years for a CGI degree just for me to prompt this from my bed The lighting on the monitor head is actually insane if you look closely Made it with Seedance 2.0

@bram_naus No it is bad and do not have 450 weight. I use Fira Code 450 retina one

Wake up. New Inter killer just dropped. 10 weights. 20 styles. Completely free. Link below.

@cursor_ai Bigger is not better, please refine what you have before starting it. It must be efficient, otherwise you already lost agains deepseek v4

Together with SpaceXAI, we’re training a significantly larger model from scratch, using 10x more total compute. With Colossus 2’s million H100-equivalents and our combined data and training techniques, we expect this to be a major leap in model capability.

Introducing Composer 2.5, our most powerful model yet. It's more intelligent, better at sustained work on long-running tasks, and more reliable at following complex instructions. For the next week, we’re doubling the included usage of the model.

@ThePrimeagen You know I could prompt LLMs in Turkish language but all they do is translate, process then translate back. As we know this is a new code language and agents did not trained with it, it basically do similar process which indeed increase token usage and drop the quality.

I have reviewed the language and really tried to understand this, but I really do not understand this language's purpose other than engineers with too much free time, free tokens, and a marketing budget. I was very excited to read about a language is "agent's first." Its just zig with a touch of java and rust...