drbscl

>Let’s be clear: if I could have a free crabchair tomorrow, I’d be out the door, in the line to pick up my new robot legs before I knew where to go.

So you're asking for a crab chair?

Jokes aside, yeah, it's expensive and likely impractical, but exploring the solution space is what this prototyping is for

You’re missing my point. I’m saying anthropic are exaggerating their results.

Not disputing the increase in quality, just stating that non-cherry-picked benchmarks show it is more verbose at Max effort

So did I. Unfortunately it's even more verbose according to https://artificialanalysis.ai/models/claude-opus-5-5#token-u...

Unfortunately, they're full of it https://artificialanalysis.ai/models/claude-opus-5-5#token-u...

It does work out to be a similar cost per task though

In these metrics, yes. In the reported training budgets of anthropic/openai, who knows?

My understanding of tech salaries in China is that they are pretty decent, but not as high as in SF; closer to typical European salaries.

Mostly due to lower cost of living; Shenzhen is way cheaper than SV

Omni usually means multimodality (in terms of input and/or output type, text, images, audio, etc)

Yes, but not commonly. Hence my request for elaboration

Yeah honestly fair point

Though I imagine it’s more about getting upfront payment for tickets to cover suppliers

The organisation has to still pay their suppliers, and have to comply with anti money laundering laws

> extremely right wing misogynism of DefCon.

Can you elaborate? I thought a lot of LGBT people attend regularly?

Breaking TOS isn't illegal per se. It just allows for denial of services, and may define terms by which the provider can reclaim costs.

>As we develop GLM, the model sometimes exhibits capabilities that surprise us, and even unsettle us.

Come on now

Also, why would they introduce themselves on their own blog?

No, just a push for regulatory capture, nothing more

>Without the local agent, you are shuttling off all your data to be persisted by an AI cloud provider in case they need it.

I'm aware of the privacy implications. I'm talking about Apple's "private cloud". All of this data is already on iCloud anyway.

>The foundation models run on the device for access to the local graph database and to interact with local actions (intents).

This does not require an on-device model.

Yeah, the 15 owners got hosed badly by it, especially since they were marketing it as the "AI iPhone" when it launched.

Personally, new Siri is the only reason to upgrade from 13 Pro, and it's just not worth the money to bother.

I think technically it can, but the cameras are black and white, so it would be very subpar for AR out of the box.

There is some chatter about using the expansion port to add a 3rd party colour camera to get over this, but I wouldn't hold my breath for it.

Edit: I stand corrected, there is such a 3rd party camera linked on the page!

I wish they allowed iCloud subscribers with older hardware to use cloud-only Siri AI.

I imagine they wouldn't shift as much hardware that way though.

> I don't get that mentality. Millions of units, shares many parts with other models, OK margin. This is no reason to kill a product.

At 2% of sales, the opportunity cost is the reason to kill the product. Keep in mind this still needs design, manufacturing, marketing, and distribution resources.

I say all of this as a fan of the Mini series.

Glad they managed to get a way out after their financial difficulties. Hopefully they got a decent exit.

Yes, as the owner of an LG OLED, I paid a premium for a good quality panel, and expect it to be mine.

Also, I anticipated ads on a smart TV (unfortunately it's inevitable), but (wrongly) assumed that such invasive tracking and "we own the glass" would be a bar too low even for the budget manufacturers.

I'm never buying any LG product ever again.

Given that they push capabilities at the pareto frontier, yeah

A lot of us use these in our services, so we're getting an upgrade "for free"

Sold after the price rise from this announcement. They can make the sales projection, but it doesn't mean they'll hit it:

1. Small models are rapidly growing in capability, require less compute to train and serve

2. There are more suppliers now, both in China & the US (OpenAI even have their own inferencing hardware now)

3. Memory still constrains how much they can ship in the short term

Yes, that's correct

> suspiciously fast

They're reporting ~30tps, that's about in line with many medium sized models served by Chinese providers

Grift so big it has a gravitational pull

Bending Spoons' entire business model is buying businesses that are failing/non-profitable despite having customers. Is it much of a surprise that the first thing they do is to massively increase pricing to make RoI?

I don't recall her name, but a past government hired a consultant before implementing the age verification laws, who warned that this would obviously happen, and they promptly fired her.

That's not what the other commenter is saying. The other commenter is stating that by doing this logo trick, bsky is very similar to competing platforms