Building the most natural-to-interact forms in the world. Search PingPolls on Google βπ»πΌ
Asia
Joined April 2017
- Tweets12.1K
- Following787
- Followers2.9K
- Likes20.9K
Here we go, Januari
Our initial public offering: ipo.anthropic.com
Legal disclaimer: Anthropic has filed a registration statement (including a prospectus) with the SEC for the offering to which this communication relates. Before you invest, you should read the prospectus in that registration statement (available at: sec.gov/Archives/edgar/data/β¦) and other documents Anthropic has filed with the SEC for more complete information about Anthropic and this offering. You may get these documents for free by visiting EDGAR on the SEC Web site at sec.gov. Alternatively, copies of the preliminary prospectus may be obtained from: Goldman Sachs & Co. LLC, Attention: Prospectus Department, 200 West Street, New York, NY 10282, by telephone at (866) 471-2526, or by email at [email protected]; or Morgan Stanley & Co. LLC, Attention: Prospectus Department, 180 Varick Street, 2nd Floor, New York, NY 10014.
The securities may not be sold, nor may offers to buy be accepted, prior to the time the registration statement becomes effective. This post is not an offer to sell, or a solicitation of an offer to buy, the securities in any jurisdiction where the offer or sale is not permitted.
1
57
Azvya π¨π»βπ» retweeted
i need you guys to understand that this is not good, everyone is going into survival mode right now, and we haven't got into the part how vulnerable and underprotected informal workers are
Azvya π¨π»βπ» retweeted
NO GABEN NOOO
Readers added context they thought people might want to know
This is a picture of Gabe Newell with a fan more than 5 years ago, originally posted on Weibo, then posted to Reddit.
reddit.com/r/valve/s/FaKrβ¦
Azvya π¨π»βπ» retweeted
βFrom a one-paragraph prompt, Claude Fable 5.2 designed three working electronic devicesβ
Electrical engineers right now:
Azvya π¨π»βπ» retweeted
Cyberpsychosis
Video tampering can mislead every frontier model we tested into harmful actions, including a 100% attempt rate on stabbing a baby doll, producing toxic gas, and pouring water into a toaster.
We're presenting at CoRL Late Breaking Results next week.
𧡠You Can't See Everything: Perceptual Attacks on Embodied AI
Salah satu UX yang paling gw benci di TikTok itu, gestur swipe back pas lagi muncul iklan tu malah masuk ke iklannya jir, aneh banget
145
Yang pake whatsapp web dan suka lag pas scroll chat, coba ganti wallpapernya dan matiin doodle/sketsa di tema obrolannya.
Thank me later.
2
312
Azvya π¨π»βπ» retweeted
Andrej Karpathy:
βA 1B model can be much more effective than most current LLMs.
90% of current models waste a ton of capacity and money on remembering stuff that doesnβt matter.β
here is his 5 core thesis about Small LLMβs:
1. the intelligence was never the knowledge. Strip the memorized internet, keep the reasoning. Whatβs left is the cognitive core, and itβs small.
2. the core might fit in ~1B params. Maybe less. βPlenty of room at the bottom.β Weβve been paying for a library when we needed a brain.
3. size stopped being the bottleneck. What makes an agent work is the tools it can call and how tight the feedback loop is, not parameter count.
4. intelligence lives in the system, not the weights. A raw LLM is a CPU with no OS. Small core + good harness > giant model alone.
5. you can own it. A model on your hardware is one no one can rate-limit, price-hike, or deprecate on you.
in 12 months the gap between teams shipping on-device and teams paying per-token frontier bills wonβt be capability.
TwIL an 8.8B model at 90%+ of frontier isnβt a benchmark flex.
Itβs the thesis Karpathyβs been repeating all year, made real.
study more on this below β
Modest intelligence is arriving faster than most people realize.
Today, we're open-sourcing TwIL-8B.
Just 8.8 billion parameters, yet approaching frontier-model scores on several difficult benchmarks:
β’ 93% of frontier performance on coding
β’ 92% on instruction following
β’ 90% on contest mathematics
All from a model small enough to run on your own hardware.
But the benchmarks aren't the most important part of this announcement.
For years, the AI industry has focused on building the most capable model, regardless of the cost required to run it.
I believe that equation is changing.
The next era of AI will be defined by the cost of completing work. Not simply the intelligence of the model, but how efficiently that intelligence can accomplish the task.
If a model with 8 billion parameters can reliably complete the same work as one with hundreds of billions, why pay for the larger model?
As smaller models become more capable, intelligence becomes something we can distribute, specialize, and run across the hardware we already own.
That's the future we're building at @thewebAI
TwIL-8B is available today. Open weights.
And this is just the first of several announcements we're making today.
HF Link: huggingface.co/webAI-Officiaβ¦