𝔠𝔦𝔫𝔢𝔰𝔱𝔬𝔭𝔦𝔞.

@NVIDIA_AI_PC @huggingface @googlegemma @ggerganov Do you want to spend $4k on a single card, or $4k on an entire machine with more ram?

.@GoogleGemma 4 31B is up to 2.7X faster on RTX using llama.cpp. Thanks to @ggerganov for working with us to make this model fast.

@BadAINoData @krazyd0h @HuggingPapers @cyb3rops “Entry level” when the card costs $4k+ lmao

@krazyd0h @HuggingPapers @cyb3rops RTX 5090 is an entry-level consumer GPU and has 32GB VRAM

NVIDIA just released a quantized Gemma 4 31B on Hugging Face NVFP4 compression delivers 4x smaller weights with frontier-level accuracy. Runs on consumer GPUs with a 256K context window.

@anemll @Prince_Canuma @GoogleDeepMind Tell me more of this wizardry

@Prince_Canuma @cinestopia @GoogleDeepMind Now you can add a GPU to the Mac Studio, so go for both 😀

Gemma 4 by @GoogleDeepMind is here! Day-0 Apple Silicon support 🍎 Specs: 👀 All are VLMs 🧠 31B / 26B A4B for complex tasks 📱 E2B / E4B for edge (with audio support) 🔧 Base + IT checkpoints mlx-vlm & mlx-lm releases incoming... 🚀

@thestreamingdev have you tried it with hermes?

Mac Mini Gemma 4 go brrrrrr

@dali__g @Jimmy6929 @Alibaba_Qwen you're assuming people read the entire post.. lol

@Jimmy6929 @Alibaba_Qwen Reading and understanding are really two different things…

（1/8）🚀 Introducing Qwen3.6-Plus: Towards Real-World Agents! 🤖 Today, we’re thrilled to drop a major milestone in our journey toward native multimodal agents. Here is what makes Qwen3.6-Plus a game-changer： 💻 Next-level Agentic Coding: Smarter, faster execution. 👁️ Enhanced Multimodal Vision: Sharper perception & reasoning. 🏆 Top-tier Performance: Maintaining leading general capabilities. 📚 1M Context Window: Available by default via our API. Built on your invaluable feedback from the Qwen3.5 era, we’re laying a rock-solid foundation for real-world devs. Get ready to experience truly transformative ✨ Vibe Coding ✨. Huge thanks to our community! Go try it out and show us what you can build. 👇 Chat: chat.qwen.ai API: modelstudio.console.alibabacloud.com/ap-southeast-1… Blog: qwen.ai/blog?id=qwen3.6 🔔Noted：More Qwen3.6 models to come and be open-sourced! Stay tuned~ 👀#Qwen #AI #AgenticCoding #VibeCoding #Agents

@aref_vc @ollama wish image and video generation were usable on mac.. with these ollama updates it'd be an absolute no brainer.

wow ! this is awesome freaking fast ! well done @ollama so fast you can barely see what is happening, this is not even accelerated !

@CuriousRefuge seedance? what are we lookin at here

This AI parody of a dating reality TV show might even be more entertaining than the real thing. Adam Palmer nailed it, from the set and character design (that jawline 👀) to the twist that had us going “of cooourse he is” and down to the details… is that a GPU ring?!!! 😂

@BlackLabelAdvsr brother if your toilet is spraying water in your asshole after you shit i promise you maintaining your yard isn't a big deal lol

Owning a home is a money pit. If someone tells you otherwise, they’re lying to you. I’ve lost track of how much I’ve invested in my yard, AC system, air filters, plumbing repairs, water filtration, bidet filters, mold repair, painting, fence replacement, etc. Be warned!

@rushicrypto combination of getting older, and the onslaught of attention-grabbing content being forced in our faces daily. the best way to slow down time is to do things that shake your routines.

No funny shit… does this world not feel off to anyone else? Days are flying. Weeks gone in a blink. Whole months just disappearing. I swear it used to feel slower… like you could actually sit in a moment. Now everything feels rushed, like we’re just being pushed forward nonstop.

@kyle_c_52 @lovs2splooge @RomanNumeral_IV @SatisfactoryAF every day is a waking nightmare for this guy, i assure you.

@lovs2splooge @RomanNumeral_IV @SatisfactoryAF Just enjoy what you enjoy bro who cares “the side” the creators are on. Just making your life less enjoyable

Simply put: Men are underrepresented in games. We're changing that.

@lovs2splooge @RomanNumeral_IV @SatisfactoryAF you cancel nerds are so fucking pathetic. 2016 era crybaby bullshit. what's next, gonna complain about wearing a mask in public?

@RomanNumeral_IV @SatisfactoryAF Like you retards didn't spend the last decade cancelling everyone. Fuck off, y'all get no quarter. Go woke, go broke. That includes you too satisfactory (even though I liked the game)

@ContraQuintis @umyaznemo lmao guys is hating pedos an astroturfed leftist critique?

@umyaznemo Fuck you and your astroturfed leftist "critique". When we are through with one group of traitors, the commies are next.

This is America!

@thematrixb0t this has a concerning lack of mentioning trump directly. he's knee deep in ALL of this.

Guess who owns the UFC

@__tinygrad__ @SchwabsinEurope @ml0_1337

@SchwabsinEurope @ml0_1337 In the metaphor I tried to keep things to scale. You also have to consider the speed limit on the road and how many stop signs there are. In this case, no stop signs, and speed limit of 40 or 120 depending on how much your town spent on road.

If you have a Thunderbolt or USB4 eGPU and a Mac, today is the day you've been waiting for! Apple finally approved our driver for both AMD and NVIDIA. It's so easy to install now a Qwen could do it, then it can run that Qwen...

@rj @LottoLabs lol for real so many 3090s listed for $500...

@LottoLabs eBay has so many scammers, im scared but want one.

You are required to own at least 1x3090

@GamewithDave Arcades.

People who actually experienced the 1990s: What is something you miss from that decade that just isn't the same today

@0xSero @neuralSWE @__tinygrad__ Wouldn’t you run into the same vram bottlenecks though?

@neuralSWE @__tinygrad__ How much vram is that? I'm thinking as low as 2k usd

The first company to make AI boxes, with specialised AI models trained to fit on that hardware will be the next Apple. Would you buy? Should I start a company doing this?

@imjszhang @ollama lmao "table stakes" becoming the new em dash.

Ollama is now updated to run the fastest on Apple silicon, powered by MLX, Apple's machine learning framework. This change unlocks much faster performance to accelerate demanding work on macOS: - Personal assistants like OpenClaw - Coding agents like Claude Code, OpenCode, or Codex

@jdros15 @AndrewHart @trq212 that's the kicker. my tinfoil hat take is the cloud providers have inflated the prices of GPUs to such an extent that their services are the "only" option. could just be natural market adjustments, but it's at least a friendly convenience for the SaaS models.

@cinestopia @AndrewHart @trq212 if only we have a good enough model that doesn't require a damn 5090 to run.

To manage growing demand for Claude we're adjusting our 5 hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged. During weekdays between 5am–11am PT / 1pm–7pm GMT, you'll move through your 5-hour session limits faster than before.