eshack94

Oddly, it seems like it hasn't been touched in 6 months (and last release 7 months ago). I wonder if it's been abandoned, or simply deprioritized in favor of other work.

I digress, but GitHub's GraphQL API has been neglected for years. Or at the minimum, they have not maintained feature parity with their REST API, so there are a good number of cases where you simply cannot do what you need to do without using their REST API.

Would you say that it's ready for prime-time on a developer machine?

Should developers upgrade yet, or wait?

Can this be disabled?

Asking the real questions. I've been wondering what the holdup on that is.

Does anyone reading this have additional knowledge or insight on this?

It appears that the only available endpoint (as of this writing) requires enabling "Allow paid endpoints that train on request data" in the OpenRouter privacy settings. I hope additional paid providers will become available that don't require training on data.

This blog post illustrates the importance of NEVER outsourcing your critical thinking or outsourcing decision-making to an LLM. And to never take shortcuts with learning. Learning is hard, but learning things properly allows you to understand what is happening and allows you to ask the right questions about whether the changes an agent wants to make are the changes that will best serve the goals of the project (without creating an unwieldy level of tech debt in the future).

Personally speaking (and I'd love to hear others' takes on this): when using an LLM for work-related and development tasks, I never use the "full-auto" mode, and I never manually approve of something that I don't understand. When I don't understand something an agent wants to do, I go on a side-quest to learn more about said thing and to educate myself first. This takes extra time, but I feel that it's the right thing to do, so I can at least approve/deny/redirect from a more informed position, rather than flying blind and hoping for the best.

In addition to what the author discusses, I think skill-atrophy, stagnation due to complacency (i.e.: "why grow and learn if an agent can do it" mindset), and cognitive laziness are additional risks that come with overrelying on LLMs. Humans were meant to think. LLMs are a tool.

"chocolate product", these days, or so it seems.

I've noticed this trend of the sampler knobs being removed. Can you explain why this might be the case?

That's exactly right.

The Claude Pro subscription is basically useless at this point, in terms of usage limits with respect to the settings required to achieve actual useful output.

Agreed, they also have great documentation. There's something to be said for documentation that is so concise, well laid out, and immediately actionable for those looking to get started quickly.

At this point, I feel like there should be a HN post whenever there ISN'T an issue with some GitHub service. Otherwise, it's business as usual...

I appreciate your guidance here. Thank you very much. I will start doing this.

I'm in the same boat as you. Wish I had known this before my subscription renewed. There's no longer any value in paying them for this service when I can cut them out of the equation and pay the model providers directly.

I've been thinking of doing this — using one of the "pretty good but not Opus 4.6-good, YET very cheap" models for the implementation part of more basic code features, AFTER first using Opus 4.6 high for the planning stage.

Do you think this would be a decent approach?

Also, which client would I use for this? OpenCode? I don't think Claude Code supports using other models. Thoughts?

Not saying anything about the article at-hand, but assuming we were able to detect it with such certainty, I think it would greatly increase the funding, rigor, and breadth or research devoted to finding a cure or treatment that actually worked.

Is this the final straw that kills their platform?

Is this a joke or are they serious with this?

Lol enshittification begins [continues].

It took me way too long to find that (I didn't notice your comment til after).

Sigh at the clickbait headlines.

I might be off-base here, but I get the sense that OP's dysphoria/ennui/ambivalence/Weltschmerz is less about what they're doing as an occupation and more other things that need to be explored more deeply. 27 can be a difficult age with many conflicting/confusing feelings and thoughts, and sorting some of those out might be the key. Again, could be way off-base here — impossible to know from a single short post like this.

Are there any higher-context 70B finetunes yet with good benchmarks?

As well as being a pain to search, it's complete black hole for information.

The trend of more and more projects and companies using Discord as their primary forum for support, docs, and discussion boards is probably not one that will be looked back on as having been a good idea.

I worry that there's so much valuable information stored in prior chats that will simply disappear if Discord ever goes away.

In addition to this, this valuable information is not accessible/searchable via the web without a Discord account. Plus there's the 100 server limit for free accounts, which can become a problem depending on how many projects/communities one is involved with.

That being said, Discord is convenient and cheaper to license/host communities than Slack, which I think is part of the allure.

Does anyone know (or is it publicly known) when the 400B variant will be released?

It would be awesome if all these newspaper archives were freely available via Archive.org or a similar service. Seems strange to need a subscription to view old newspapers. Perhaps part of the fees pay for hosting costs and the work that goes into compiling so many sources and slapping a easily navigable UI on top of it, etc.

BetterSnapTool is great for this. Check it out.

Angel + Zeus = God. o_O

This little game is too entertaining. I hope the author open sources it, or writes a technical blog post on how it's implemented.

This isn't universally true. Some people truly enjoy experiencing other cultures and parts of the world for its own sake. Not everyone who travels for pleasure and curiosity are doing it "for the gram". That being said, there are also many people who are.