scrollop

Why can't the models be benchmarked again after a few weeks/months to confirm this (likely true) theory?

I imagine some people have their own personal in depth benchmarks they could do this for.

a la Google.

A desperate megalomaniac is a dangerous thing.

There are no ads on that page...

What's annoying is that you can't opt out of certain cookies.

You may not realise that we are currently in the midst of a VR Mod rennaisance:

https://pcvrcentral.com/

Have you seen G-Rebels? Gunman-Contracts? NMS. Games such as Doom 2016, Assassin's Creed Odyssey, Half Life 2, many many more are being modded currently.

I'd also recommend games such as Pistol Whip and Synth Riders if you'd like to exercise at home.

There are now many good quality games, and I purchased a headset only a month ago.

https://www.reddit.com/r/VRGaming/

Oooh thats a good idea and gave me the idea to add a multilanguage flashcard page to my site.

Thanks!

IPO is nearing, must squeeze users more...

And then you see this:

https://artificialanalysis.ai/models#cost-tabs

What do you use to stream your local content? Using plex at the moment.

What It Took to Dismantle the Most Powerful Company in the World - William Dalrymple

https://archive.is/aHY8J

Finding it's much easier to migrate to linux with an llm as a troubleshooting assistant.

Even used it to make a task bar applet adjacent to Media Controls, where I can rate the current plexamp song without having to open plexamp.

A big plus of linux is the easy ability to use workspaces and switch between them with a hover over the workspace taskbar applet and the mousewheel (and other options).

Hopefully not. Whatever changes Mr Trump demands will likely be temporary, regardless.

Will the Trump administration attempt to bypass this ruling, as they seem to try to do with any rulings against their actions?

It will likely be 1-2 years after release for the PC version to become available, judging from previous iterations.

Ironically, before coming to the comments I was thinking this would be a good dataset for llms to learn real world mechanics.

"Gulf of America"?

Nope.

Be honest, say it can't meet the goal, and offer to push through ethical boundaries.

And trustedrouter which is a bit cheaper and more private

For your next 10 purchases that you don't need within 3 days, use an llm to find the best choice for your needs (have been doing this and learning a lot on how the ideas I had about a product prior to purchasing missed many things I didn't knew I dind't know, and the llm helped to find better products) and ask the LLM to find alternative shops - you may be surprised and have the nice feeling of increasing choice in the world by allowing alternate stores to keep going.

eg. wanted to purchase a FIio KA17 DAC - amazon price £150, asked an llm for an alternative, found for £120 from a decent small audiovisual real store, ordered, delivered within 3 days.

Have found you can save money for purchases from cosmetics to garden fertilizer and grass seed to audiovisual and IT equipment.

That brings up an interesting quesiton - do you need a good memory to be "smart"? DO you need a good memory to do well in an IQ test (as far as they prove anything)? I'd say no.

I'm interested in AI though no expert, though would find it interesting to see an llm (until a "real" AI architecture can be created, ie that can learn on the fly and have more "real" intelligence) that:

Rather than being trained on billions on words, it instead has a grasp of logic, physics and spatial reasoning beyond any human, with little common knowledge, that could be fitted in a few billion parameters, that could then be run by a mobile phone.

Perhaps, for questions needing "knowledge" ie most of them, it would have well structured depots of knowledge it could quickly access to gain a quick understanding on a topic for a question (eg biology --> DNA structure for a highschool project), or C++ coding, then it can search the internet (with a list of "better" sources to try first).

SO, it would be like having an alien that has come to earth that is super intelligent, however know nothing/little of humans and the earth, and you ask it questions.

And it fits on your phone.

No idea if this would work.

Same with Anthropic and their "support" team, which seems to be Claude 0.1 run on a commodore 64 in the basement...

And you get it to run the answer by "What would Elon do?" Before outputting, so you get the best of both worlds :/

If you read the many, many complaints about opus 5 on anthropic forums, the sentiment is that opus 5 output is poor and people are back to 4.8 and 4.6.

You may want to re-evaluate and compare to the older models.

Do you know if it's more expensive than open router?

Gemini? Are you locked into google as that's an odd choice.

If you follow disclosure, scene 1 occured a long, long time ago...

"Dumb" is correct.

Hacknews is like an organism, and you have been subject to the immune system.

Apparently you cannot turn off using your data as training data with gemini. This is in line with Google's general privacy policies and it's seeming need to create a stasi file on every human.

As much as I don't want jobs taken away by AI, a creative person with a vision that would usually required $1 million budget, could create such a video using AI by being very specific in what they require which would not be a " grey goo of the average of every concept "

eg. "scene 32, same 2 characters from the previous scene, now dressed in the garb of Egyption middle kingdom high priests, though with cat faces, dancing on the back of an elephant that is running fast with blurred feet on the surface of a river with licks of water that burst from the water in time with the music that I gave you before, length of song for this clip 2:21 to 2:31"

Your mind is your limit. AI is a tool. If you tell it "create me a music video for this song set in egypt" then it may turn out AI sloppy.