Art9681

6 ms·

91.
by Art9681 10mo ago
Truth. It's a hard pill to swallow. But this is the reality. It's irrelevant if it has our approval or not.
92.
by Art9681 10mo ago
SynthID can be easily circumvented. One of the top posts in a certain subreddit today relates to the topic for those willing to look.
93.
by Art9681 10mo ago
We have different definitions of sovereign state apparently. "In his time in office, Venezuelan President Nicolás Maduro has stolen two presidential elections, electoral monitors and human rights groups contend, while jailing critics a
94.
by Art9681 10mo ago
At this point they might as well implicate everyone who contributed to computer science as being at least partially responsible, willingly or not, on creating the monster. What were we trying to do this entire time? Automate right? Get to t
95.
by Art9681 10mo ago
Thank you. This is a perfect example of clickbait. I trusted the HN crowd, clicked the link, and immediately realized the trap. I'm upset at how effective it is. And also commend the author for publishing an article specifically engine
96.
by Art9681 1y ago
Well said. No organism willingly commits perceived suicide unless it's a viable strategy for its continued existence. The reason a thing exists, is because it hasn't tempted its potential predator.
97.
by Art9681 1y ago
Best we will ever do is create a model of intelligence that meets some universal criteria for "good enough", but it will most certainly, never be an objective definition of intelligence since it is impossible to measure the system
98.
by Art9681 1y ago
Bummer. Back to Docker then.
99.
by Art9681 1y ago
It shifts your cognitive tasks to other things. Like every tool. The tool itself is an abstraction over tedium. We built it for a reason. You will spend less time thinking about some things, and more time thinking about others. In that rega
100.
by Art9681 1y ago
Depends on the model. gpt-oss-120b will easily crunch large prompts in a few seconds. It's remarkable. It's gpt-4-mini at home.
101.
by Art9681 1y ago
It can be summarized as "Did you RTFM?". One shouldn't expect optimal results if the time and effort wasn't invested in learning the tool, any tool. LLMs are no different. GPT-5 isn't one model, it's 6: gpt-5,
102.
by Art9681 1y ago
https://www.reddit.com/r/MyBoyfriendIsAI/ Way down the rabbit hole we go...
103.
by Art9681 1y ago
The thing is that someone has to clean the toilets, sweep the floors, maintain that public transportation, or have the skills to fix up your car, etc. Dignity means a lot of things to different people. If you live in a society that enables
104.
by Art9681 1y ago
None of the major players have ever been quiet. DeepSeek enjoyed about a week or two's worth of press before its spotlight was stolent from the next great model. It never held the top spot, ever, mind you. So I don't understand wh
105.
by Art9681 1y ago
It has everything to do with the way you use it. And the biggest difference is how fast the model/service can process context. Everything is context. It's the difference between you iterating on an LLM boosted goal for an hour vs
106.
by Art9681 1y ago
Does the Central Bank of Brazil publish data on the cost of running the network and how it is funded? Or is there a hidden tax that eventually makes its way to the customer in order to keep it going? Nothing is totally free. Who pays for th
107.
by Art9681 1y ago
This is exactly my same take. Any tool an engineer is inexperienced with will slow them down. AI is no different.
108.
by Art9681 1y ago
Any modern MacBook can run those tests 100x faster than the crappy cloud runners most companies use. You can also configure runners that run locally and get the benefit of those speed gains. So all of this is really a business and technical
109.
by Art9681 1y ago
It does not. Context is context no matter how you process it. You can configure tools without MCP or with it. No matter. You still have to provide that as context to an LLM.
110.
by Art9681 1y ago
It wouldn't. There is a difference between theory and practicality. Just because we could, doesnt mean we should, especially when costs per token are considered. Capability and scale are often at odds.
111.
by Art9681 1y ago
RoPE scaling is not an ideal solution since all LLMs in general start degrading at around 8k. You also have the problem of cost by yolo'ing long context per task turn even if the LLM were capable of crunching 1M tokens. If you self hos
112.
by Art9681 1y ago
The benchmarks are not reflective of real world use case. This is why OpenAI dominates B2B. As a business, its in your best interest to save money without sacrificing quality. "Follow the money." Businesses are pouring money into
113.
by Art9681 1y ago
Better how? DeepSeek has never held the top spot in any aggregated benchmark. The Chinese bot armies are certainly better at convincing the internet they are trailing western models, despite the fact that in practical use this is not the ca
114.
by Art9681 1y ago
This is completely irrelevant without knowing if you are effectively prompting each model. Your workflow may just be suitable for a particular model and not others. And tuning a workflow for each model is tedious. I seriously doubt there is
115.
by Art9681 1y ago
Distilling western SOTA models that now summarize their thought process is expensive in 2025.
116.
by Art9681 1y ago
"The hardware is the easiest part." - AMD probably
117.
by Art9681 1y ago
A lemon is on-par with the best western models for the majority of use cases because they do not require "state of the art" intelligence to solve or respond to the user's query. This is what the benchmarks show. For anything
118.
by Art9681 1y ago
What's the point of stating the obvious if the obvious won't change anything? Things evolve. Winners win and losers lose. Change is constant. And? Does that somehow mean there's nothing to see here and we should move on?
119.
by Art9681 1y ago
This is the wildest headline i've read in quite some time.
120.
by Art9681 1y ago
This is a "BlockBuster laughs Netflix out of the room" moment. I am a huge fan of QEMU and used it throughout my career. The maintainers have every right to govern their project as they see fit. But this is a lot of mental gymnast

More ›