Wondering where everyone will be tomorrow?😌
This is it. The biggest conversations, the best vibes, and unforgettable moments all in one place.
📍ThisDay Dome, Kur Mohammed Ave, CBD, Abuja 🕛 12 PM–6 AM 🤗
Register here : https://t.co/kT1XylmyJh
#TheGatheringOn100#LiveIt100
🇺🇸 BlackRock CEO Larry Fink remains bullish on markets over the next 12 months.
Despite Bitcoin correcting from $126K to around $60K, Fink sees further upside ahead.
When the world's largest asset manager speaks, markets pay attention.
#BTC#Bitcoin#BlackRock
Fanum explains why he has returned home from Streamer University 2 , says there’s some issues with Kai Cenat behind the scenes and that things will be explained later on 😬
27B-class intelligence. 7 GB. Running on your Mac.
Ternary Bonsai 27B (@PrismML) is squeezed down to true 2-bit — where most sub-4-bit models fall apart, this one
still reasons, writes code, and runs agents. ~95% of FP16 intelligence, fully on-device.
🧠 80.49 / 15 benchmarks · 💻 Code 86.0 · 🛠️ Agentic 74.0
$ rapid-mlx serve bonsai-27b-2bit
https://t.co/5kZHwYgUp1
Seeing Hy3 at the top of @OpenRouter is incredible. 🏆 I’ve been relying on it personally, and it’s truly a game-changer for local LLM workflows.
We’ve made sure it’s fully supported by rapid-mlx so you can experience it too. Worth a look! ✨
Great work @andyfang. This marks a profound shift in inference—moving from humans querying AI, to AI orchestrating human action in the physical world. Starting with delivery, expanding to an entirely new agentic economy.
Duke Dennis explains why everyone in AMP is donating 95% of their streaming income , says it's to help Kai Cenat with the Streaming University 2 costs since Kai only makes $2 million a month 😬
This might be one of those "wait… this runs on MY Mac?" moments. 🤯
rapid-mlx now serves two open models that have no business running on your own hardware:
🐘 HY3 295B by @TencentHunyuan — a frontier brain going toe-to-toe with Claude Opus 4.8. Fully local. Nothing leaves your machine.
🐜 Bonsai 1.7B Ternary by @PrismML — 0.47 GB. Half a gigabyte. Landing in GPT-4o's neighborhood while streaming 300+ tok/s on Apple Silicon.
$ pip install rapid-mlx
$ rapid-mlx serve hy3-preview-4bit
$ rapid-mlx serve bonsai-1.7b-2bit
→ https://t.co/5kZHwYgUp1
Your local LLM is thinking things it never tells you.
@AnthropicAI just revealed the hidden "workspace" where this happens. We've brought their J-lens technique to your local models with a single command:
rapid-mlx jlens "why is the sky blue"
Point it at any open model on your Mac and watch it form its answer—layer by layer—before it types a single word. 🧠🔮
📄 https://t.co/oMmEfjhGor
⭐ https://t.co/MAWcFRVeJT
📉 Trader hat on: I agree with @blknoiz06. AI is perhaps ~80% its cycle peak, while crypto is currently underwater, hated, or ignored.
🏗️ Builder hat on: AI and world models are infinitely more exciting to build in. They carry the real weight of human technological advancement. 🚀 (Smart contracts have been here for a decade—where is the next tangible breakthrough in crypto for real people?)
Ive been saying for last week now in my telegram channel BTC and Alt coins are bottomed we are on our way into a new bull run now.
It’s time to buy big in spot and Long your longs BTC $57k was the bottom everyone calling for $40k are noobs.
Generational wealth is incoming.
Fanum explains why he won't be attending Streamer University 2 anymore , says Kai Cenat requested he pays 90% of his earnings while he streams at the event and that he just couldn't do it 😬