sinuhe69

4 ms·

1.
by sinuhe69 3d ago
If it works for you, great! Just keep going. I only have the nagging question that if it’s so easy, why would the clients not do it by themselves? Certainly they understand their problems better than you and they can adapt to rising issues
2.
by sinuhe69 5d ago
Absolutely! I’ve never lied to anybody and any work we copied was purely accidental.
3.
by sinuhe69 5d ago
Snap is designed to be more expressive and powerful than Scratch. But I find debugging them is very painful. Changing the name of a variable or block for example, could create “holes” in the calling sites but the system can or cannot repor
4.
by sinuhe69 5d ago
"OpenRouter runs per-provider benchmarks on the same model: GPQA Diamond and TAU-Bench Airline (a tool-calling task)." why I haven't never seen it? Click on the link brings nothing! Benchmark is only for the model. Per provid
5.
by sinuhe69 6d ago
Then I guess you will need millions tones of perfect mass to light converter like Astrophage just to get to the neighbor. The universe is simply too big to comprehend.
6.
by sinuhe69 6d ago
Your benchmark is a curious one. I didn't see you included Muse Spark 1.3 contributor even though its price is much lower even than DeepSeek. The low price changes many recommendations completely. And FWIW, DeepSeek retain and train on
7.
by sinuhe69 6d ago
While I also have my reservation against Elon and his activities, I can read these posts on my browser without any account.
8.
by sinuhe69 6d ago
No, if they want they can easily compare the strings verbatim because these exact phrases are so extremely rare that it almost certainly isn’t in other conversations. But of course they wouldn’t do it. Why would they?
9.
by sinuhe69 6d ago
I guess someone can twit Hassabis and ask him :D He certainly has access.
10.
by sinuhe69 6d ago
More parameters = more facts stored. Knowledges are almost incompressible, where strong reasoning only requires a 3B core or so.
11.
by sinuhe69 8d ago
No wonder - they're all working for ant! Birds of a feather flock together.
12.
by sinuhe69 8d ago
I'm tired of this, but please read the post of Tao. It’s listed in the top comment of this thread.
13.
by sinuhe69 8d ago
More like helped improve our work (the disproof)
14.
by sinuhe69 8d ago
Skills are for me special knowledges that an agent in general doesn’t know in details: using a specific framework, create reports, new programming languages etc. So having them not only enable the AI to work more effectively but also more
15.
by sinuhe69 9d ago
When I work on a new problem and the agent struggles with it, I will ask the agent to distill the knowledge and experiences it gained during the session into a skill file. Then I review and publish it depends on the assistant system. I fin
16.
by sinuhe69 11d ago
Congratulations! An immensely important achievement, I’d say. IMO, this success has a great implication for accessibility and sovereignty of the EU. Although ESA can launch from French Guiana, the spaceport is an ocean-away from Europe! Tha
17.
by sinuhe69 11d ago
No. It’s simply about accessibility and sovereignty. Even though the ESA can launch from French Guiana, the space port is an ocean-away from Europe! How much do you think that fact will affect the cost and the launch cadence for Europe? Now
18.
by sinuhe69 14d ago
No, they use geo-fencing and redirect the request to the FBI site! :D
19.
by sinuhe69 14d ago
When I accessed the site, it showed the FBI badge says that the site was blocked and redirect to fbi.gov !! WTH?
20.
by sinuhe69 20d ago
If AI and robots are going to replace people in such a scale, without creating new jobs, who will they sell the products or services to? I cannot believe Bill Gates has not even conceptually engaged with that question. That said, I share th
21.
by sinuhe69 22d ago
22 times more efficient is not like 22 times more powerful. It’s extremely harder to close the gap in power efficiency than in raw power. Simply because the power efficiency we see is the result of billion years evolution optimization. But
22.
by sinuhe69 23d ago
Reminds me of the story of Million Dollar (Pixel) Map. It's a piece of internet history! https://milliondollarhomepage.com/
23.
by sinuhe69 23d ago
I would find the news much more interesting if they put the new processor in a laptop with 48 GB DDR5 unified memory or more! Right now, M5 with its neural execution engines and unified memory architecture present a unique local AI computin
24.
by sinuhe69 26d ago
Me too. It demanded deep digging to find out how it was done and how is it different from other products. And my brain was resisting all the way through! I wish the author had made a mental image like “OzBrain is a Zettelkasten system built
25.
by sinuhe69 26d ago
I think the central question for a such memory system is whether we or the agents can find the relevant information and how to organize these data as changes continues to come in. Would we miss something in the retrieval process? How do we
26.
by sinuhe69 1mo ago
Any implication for the privacy and confidentiality of our queries? Openrouter yes sees everything. But as long as they are independent and establishing their core business on privacy and confidentiality reputation, I feel more comfortable
27.
by sinuhe69 1mo ago
Very exciting experiment! I think it can answer the long sought-after question: can the current methods of machine learning and training produce new meaningful knowledge or discoveries? It’s a good test bed because the curated curriculum is
28.
by sinuhe69 1mo ago
Not just domestic. They come from other Asian countries as well. Asia is huge with billions of people and their time zones are not so much different from each other.
29.
by sinuhe69 1mo ago
I read the same. Refreshingly honest, straightforward and many useful information included. It is a breath of fresh air.
30.
by sinuhe69 1mo ago
Now GLM 5.3! And they explicitly confirms my theory: "Scaling post-training is all we did for GLM-5.3. With GLM-5.2 we built the stack: IndexShare for efficient long-context processing, SAO for RL on long-horizon tasks, and slime for l

More ›