511–520 of 740 posts
Fable is not a tool for the average user. It’s a professional tool for highly complex work. I would compare it to a extremely high end $15k PC, or an expensive pro-grade video camera, or a freight train, or a … I would say at least 95% of the global population will not encounter a situation once in their life where it would be actually useful/warranted.
I hate to be brusque but this is cope. GPT 5.6 Sol is just as good and cheaper
Where Anthropic f'ed up was treating their monetization the way they treat model training. Turns out that success in experimentation is not transferrable. They have tried to find the highest that the market pays for sota models; however, on the consumer side, this is just too confusing and unsettling: "You can only use Fable for a week as a part of your plan" "Be ready! You have to start paying per token!" "Nevermind…
Also that the behavior of the models change, suddenly more fluff in the comments etc is annoying, but that is probably being part of using cutting edge tech.
so whats the best and most efficient coding harness and against which model? what are folks doing to keep costs low? I spent $1000 just this weekend on my personal projects for sota Claude but I feel like I can probably get much more juice if I start looking elsewhere. and resources or tech stack tips from HN?
I've been running Qwen27B Q6 with 130k context at 30tps. It's not bad.
I think with an "autopilot" mode in pi, and sub agents to handle context window management better, I'd envision you could get close to unattended workflows. I haven't quite gotten that far though.
Negative is you have to do it all, and the temptation to tinker is real.
Earlier quoted context omitted.
> How do you know there wasn't account sharing though? Even if it was, it should be able to be sorted, maybe pay some overcharge or explain, and have your fucking business access re-instated.
This is such a tough problem. Anthropic would need access to some kind of technology that could, like, intelligently handle unforeseen circumstances and nuances. Yeah, that’s definitely not something we should expect of them.
Earlier quoted context omitted.
I'm surprised no one mentions about their recent privacy violation(s). The breaking point for me was the privacy violation. They've been fingerprinting every request and violating users' privacy hoping no one would notice. Too bad, someone found out and that was the day when I cancelled my subscription. https://thereallo.dev/blog/claude-code-prompt-steganography
Like many things that "nobody's talking about", people really are talking about it. Discussion from two months ago: https://news.ycombinator.com/item?id=48734373
Its in my opinion not wide spread at all and as today is literally the first time i ever hear anybody mention this.
Earlier quoted context omitted.
remember 4 year ago we use to : have stack overflow open, documentation, obscure forums plus other tabs. An ide open with 20 tabs open each file a component, a class or an interface We also use to hold entire codebases in our brain.
And now I'm getting 10-20x as much done. I'd say the trade off is worth it. I'm struggling to scale myself even further. This tech is unreal and I have so many things I can do. For the first time, tech feels like the 90's-00's again. Everything is greenfield and exciting and big tech is struggling to figure out what to do about it. People are just hacking all kinds of stuff, and it's awesome. Feels like techno utopia…
It feels like the opposite of 90s - 00s: they were filled with periods where a person could self-study technology and get a job using those skills that few others had.
Where we are going (according to the AI-proponents) is children being able to replace you.
In brief; the 90s - 00s were a skill-valuation time, now we are looking at a skill devaluation time.
Unless you meant to say "Just like how any kid who could write broken HTML t put up a webpage could pretend to be a skilled professional, that's where we are now"...
Reads like: brilliant Carnegie Mellon University computer science grad struggles to find job where he is not replaced by cheap, inferior Indian labor that still gets the job done, even if it takes marginally longer. No shit we're all paying 清冲 Flash to do the grunt work. Turns out though, paying 清冲 Flash a few more cents does exactly what Ivy Wasp Pro Mythical does. Crazy how that works.
Earlier quoted context omitted.
Maybe not Anthropic, but LLM translations has a value counted in the trillions of dollars easily.
I see it's important but trillions? How did you quantify that number?