


fiddling with AI @futuredapps



Seems like kyutai really cooked here

It's interesting that people think prompts are the most valuable, not the infrastructure, technology, UX, internal benchmarks, ability to integrate new models quickly, or quickly assessing if adding 200$ tips to LLMs improves performance.
Billions worth of system prompts are exposed in a single repo. The scale of IP leaks is remarkable in the LLM era.

"10M context window" RAG is dead post in 3, 2, 1 ...
Introducing our first set of Llama 4 models! We’ve been hard at work doing a complete re-design of the Llama series. I’m so excited to share it with the world today and mark another major milestone for the Llama herd as we release the *first* open source models in the Llama 4 collection 🦙. Here are some highlights: 📌 The Llama series have been re-designed to use state of the art mixture-of-experts (MoE) architecture and natively trained with multimodality. We’re dropping Llama 4 Scout & Llama 4 Maverick, and previewing Llama 4 Behemoth. 📌 Llama 4 Scout is highest performing small model with 17B activated parameters with 16 experts. It’s crazy fast, natively multimodal, and very smart. It achieves an industry leading 10M+ token context window and can also run on a single GPU! 📌 Llama 4 Maverick is the best multimodal model in its class, beating GPT-4o and Gemini 2.0 Flash across a broad range of widely reported benchmarks, while achieving comparable results to the new DeepSeek v3 on reasoning and coding – at less than half the active parameters. It offers a best-in-class performance to cost ratio with an experimental chat version scoring ELO of 1417 on LMArena. It can also run on a single host! 📌 Previewing Llama 4 Behemoth, our most powerful model yet and among the world’s smartest LLMs. Llama 4 Behemoth outperforms GPT4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on several STEM benchmarks. Llama 4 Behemoth is still training, and we’re excited to share more details about it even while it’s still in flight. A big thanks to all of our launch partners (full list in blog) for helping us bring Llama 4 to developers everywhere including @huggingface, @togethercompute, @SnowflakeDB, @ollama, @databricks and many others👏 This is just the start, we have more models coming and the team is really cooking – look out for Llama 4 Reasoning 😉 A few weeks ago, we celebrated Llama being downloaded over 1 billion times. Llama 4 demonstrates our long-term commitment to open source AI, the entire open source AI community, and our unwavering belief that open systems will produce the best small, mid-size and soon frontier models. Llama would be nothing without the global open source AI community & we are so ready to begin this next chapter with you. 🦙 Read more about the release here: llama.com/, and try it in our products today.

Help me test whether my bank or Vercel will call me first. onemillionpixels.online/

Folks, do you think hosting this on Vercel and sending 4MB real-time updates to every player was a good idea? #vibejam


Seems like increasing ammo from 100 to 5,000 allowed people to go wild #vibejam

The game I vibecoded for fun was used in a class to teach kids about pixels, images, and computers. Sometimes things take unexpected turns. They called it Minecraft with painting. Glad it was before NSFW content appeared 🙈 #vibejam Try it at onemillionpixels.online

Pro tip: Check your prompt for typos so you don’t write ghoblify instead of ghiblify 🧟♀️

Another day, another small win: First bigger way of players arrived today #vibejam and somebody left a message Try a collaborative canvas painting game here: onemillionpixels . online

Take a rest from ghibli and let's play some tic-tac-toe. Green already lost this one! #vibejam onemillionpixels.online

But if you are interested: onemillionpixels.online/

How can one promote their #vibejam collaborative canvas painting game when their entire X timeline is just Studio Ghibli images?

Game is available here: - best on desktop (mobile kind of works, but - for refill catch some buckets onemillionpixels.online/