wg0

Can I put it as Air Traffic Controller? With similar error rates as humans?

That would be the litmus test.

"Does not hallucinate" is not the same as "is never wrong".

So the ATC test could be the benchmark.

I am concerned that what will happen to detailed technical write ups that defined much of the "technical internet space"? There used to be "CodeProject" (still is around probably) that used to have great articles on C, Windows Programming, Windows System programming and what not. Similar other sources on Web as well such as some TutPlus network.

Now with LLMs, they won't gain attention and all knowledge would be stuck in LLMs but there would be no knowledge base to train the LLMs on so... would this all collapse in a decade or two?

PS: Not anti AI rant, just a genuine open question.

Has anyone tried it with other models.?

Then slow down your bots too please. I deployed a website last night. 10 out of 11 GB of bandwidth went to Claude bots alone with more than 90k requests.

Hypocrites hyping for the IPO.

PS: True story.

What I mean is that price has been effectively halved.

During off-peak hours, the unit price is reduced from $0.007 for input cache hits to $0.003, $0.22 for input cache misses to $0.15, and $0.12 for output to $0.6

DeepSeek v4 Flash with high is already a really great work horse. Reliable. But this time, not only that it is better but they are reducing the price by 50% so that's great.

I also find the DeepSeek models to be more precise than Claude models (last I used 4.7) in that I yet had not the occasion where model did something unintentional that I did not direct it to.

EDIT: Updated percentage reduction.

Not happening.

Would love to see what was it.

Man - What exactly are these people?

Don't know about the bench marks but I am getting Opus 4.7 level performance at fraction of cost with DeepSeek V4 Flash set to high. It is a reliable workhorse.

I am obsessed with the idea of per tenant databases. But I am afraid of migrations. Has anyone tried that?

Basically:

"We are in favour of 3D printers but there should be a body that tests and certifies that a 3D printer cannot print anything that can be used as weapon. Anything pointy or with a spring and recoil or... or..."

> tied to the corrupt Trump administration that are neither the highest quality nor the cheapest

Someone is calling corrupt as corrupt. Surprising.

Is it the bond or is it the share price that also tanked below IPO level?

Much needed. I have hard time understanding bash.

Everyone is declaring GLM 5.2 as something that's really a big deal.

I don't know about that but based on my own experience with Deepseek v4 Lite alone (with high effort) I have no doubt in my mind that anyone claiming such great things about GLM 5.2 must be true because Deepseek v4 already is really awesome.

This administration has damaged the US soft power. For decades, US was a reliable trade partner. Not any more.

EU is looking and charting its course already. Yeah, we can joke about it, we can mock it but it is in momentum already, one step at a time.

That's why I pay them. Regularly. Without fail. Despite my token usage isn't that much.

But I vote for these heroes with my wallet. Just yesterday did again.

It's just like web scraping is impossible to guard against.

Change my mind.

Try firma.de

I hope bun desktop is coming soon?

You might dismiss it as nothing but the Linux analogy does not work here either. It is more than that and direct threat to commercial AI labs and their business model. These labs are milking bunch of foundational papers for years now and the end is near.

Going forward would be such open source, open data and open recipe models possibly someday even with the training being crowd sourced if not inference like the BitTorrent model.

Lastly, even Chinese models (GLM, Deepseek, MiMax) work really really good and any user would testify that they do not miss OpenAI/Anthropic/Gemini at all if they're using those Chinese models which is argument enough that with such models, no one is going to miss Chinese models as well.

I think open source is dead. Basic issue is - if your product is open source or even open core, building a business around it would be impossible because someone else would point an AI agent at it and would have similar thing to offer.

Hence, closed source is what's next probably. Unfortunately.

Yeah insane that people think it'll be okay in the long run but wondering how much different the financial status of other such company would be? Not much I guess.

>As for the claim coming out of Shenzen, it carries no benchmarks, gives no figure for how long the run took, how it compared to the same job on Nvidia hardware, or how efficiently the 1,000-chip cluster was used. It’s ultimately just another addition to a series of dubious claims that have come from the Chinese state without anything to back them up; DeepSeek itself hasn’t commented.

As if OpenAI and Anthropic are giving us ball to ball commentary on how their training runs go. Deepseek did train it on domestic hardware, model might be out in public soon (open weights or not) and then anyone can see what is it about.

Snake oil. Tell me something that has hard irrefutable and reproducible evidence.

Flip the rules put a blanket ban on US digital services and see what comes out of Europe within couple of years.

The only problem is - when US services are available, there's no incentive to bring anything to the market.

I would not compare to lottery because a lottery ticket is bought at pure chance whereas these startup are taken onboard after very thorough audit.

Does that work always without fail with anyone anywhere anytime?

Not specific to Opus but yes it would make mistakes. I usually try to keep context window under 10%