thestreamingdev()

all things ai and coding while streaming, DM for consulting.
founder of

[thestreamingdev() @thestreamingdev]

I ran a 35-billion parameter AI agent on a $600 Mac mini.
Specs: M4 Mac-Mini 16GB RAM

The model doesn't fit in RAM. It pages from the SSD at 30 tokens/second.

On NVIDIA, the same paging gives you 1.6 tok/s. Apple Silicon gives you 30. That's 18.6x faster.

No cloud. No API keys. $0/month.

Here's what it can do 🧵

[thestreamingdev() @thestreamingdev]

@naval @rauchg @maxhodak_ @bscholl Naval check out my startup we are turning clipping into shareable workspaces - anyone can collaborate and make clips even clip their own favorite moments!
https://t.co/f68SrY9Pc3

[thestreamingdev() @thestreamingdev]

my gf just told me gpt 5.5 is her new favorite, claude must be slacking

[thestreamingdev() @thestreamingdev]

@scaling01 They really are the worst

[thestreamingdev() @thestreamingdev]

I wonder if they’ve ever heard of autoresearch

[Anthropic @AnthropicAI]

Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor.

It’s happening faster than we thought, and the implications deserve greater attention. https://t.co/OVVPJO7VQx

[thestreamingdev() @thestreamingdev]

@zeke Savagery

[thestreamingdev() @thestreamingdev]

🥹 it’s like they’re watching our post and making solutions

[Google Gemma @googlegemma]

💻 Laptop Ready: Small enough to run locally with just 16GB of VRAM or unified memory.

[thestreamingdev() @thestreamingdev]

yummy

[Google Gemma @googlegemma]

Meet Gemma 4 12B!

A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license.

Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

[thestreamingdev() @thestreamingdev]

@zebassembly its sharding all the way down

[thestreamingdev() @thestreamingdev]

your agent can forward you links of artifacts and you can share with other agents and people. which means everyone can be on it in real time

[thestreamingdev() @thestreamingdev]

the impossible with hardware is only just starting

[SemiAnalysis @SemiAnalysis_]

Cerebras did what the industry calls impossible: turned an entire 46,225mm² wafer into one chip. Defects on silicon that big are inevitable, so they built in redundancy and custom per-batch masks that route around every bad core, landing near 100% usable wafers. The results: 900,000 cores and 44GB of SRAM on a single piece of silicon, no packaging, no off-chip hops. And they're not stopping there, now exploring hybrid bonding a DRAM wafer on top for even more fast memory. (1/4) 🧵

[thestreamingdev() @thestreamingdev]

@chatsidhartha @Cloudflare @OpenAI 😁 I beat OpenAI too it haha congrats 🎉

[thestreamingdev() @thestreamingdev]

your agent can forward you links of artifacts and you can share with other agents and people. which means everyone can be on it in real time

[thestreamingdev() @thestreamingdev]

baffling the number of companies that are merely just prompts and VCs fund it 😅

[Bill Lennon @blennon_]

AI can now make you a great parent.

Introducing Ollie: the world’s first AI family assistant that manages your family life better than any human.

Here’s how it works:

[thestreamingdev() @thestreamingdev]

They’re just using durable objects and a worker it’s actually insanely easy to build this

[Alex Volkov @altryne]

"Being rolled out to business and enterprise uses"

hahaha us $200/mo Pro users get the shaft 😂

[thestreamingdev() @thestreamingdev]

@benhylak Once again the devs on X are already ahead of the labs 😅

[thestreamingdev() @thestreamingdev]

your agent can forward you links of artifacts and you can share with other agents and people. which means everyone can be on it in real time

[thestreamingdev() @thestreamingdev]

@OpenAI @OpenAIDevs when do I get hired?

[Rohan Varma @TheRohanVarma]

We just launched Sites into Codex!

Software creation was always about more than writing code. Sites in Codex fundamentally gives the power of end-to-end software creation to every user, no matter their technical fluency.

These Sites are fully deployed to a URL, private to workspaces, come with authentication, can have static files, and can store dynamic data in databases.

It is in preview for business and enterprise teams and will be rolling out to all workspaces over the next day. Give it a try by typing @ Sites into Codex and ask it to build anything!

This project took a massive amount of effort across hundreds of people at OpenAI - proud that we were able to get this out and excited to see what you all build with it!

[thestreamingdev() @thestreamingdev]

@TheRohanVarma Great idea!

[thestreamingdev() @thestreamingdev]

your agent can forward you links of artifacts and you can share with other agents and people. which means everyone can be on it in real time

[thestreamingdev() @thestreamingdev]

@OpenAIDevs I beat you to it :p

[thestreamingdev() @thestreamingdev]

your agent can forward you links of artifacts and you can share with other agents and people. which means everyone can be on it in real time

[thestreamingdev() @thestreamingdev]

@TheAhmadOsman

[thestreamingdev() @thestreamingdev]

congrats @StephenCurry30

[Stephen Curry @StephenCurry30]

The partnership of a lifetime.

[thestreamingdev() @thestreamingdev]

@planetzomax @benfryc @astronaughtyart You get the movie render too - anything that involves camera work I feel unreal is better suited

[thestreamingdev() @thestreamingdev]

@planetzomax @benfryc @astronaughtyart youll be surprised
https://t.co/kj0Yq4BOxU