bavell

Didn't the guy flee the country after posting bail? Doesn't exactly scream "innocent".

Perhaps you could generate a few tokens before the entire model is downloaded, but since every token takes a potentially different "path" through an MoE model, you'd still need to wait for the entire download before getting deeper than a handful of tokens... which is not really a UX improvement imo.

Going on 10 years now for me, tried Helm a bit and yep - all I've really needed was a package.json deploy script with sed to bump the image version.

Funny, I've been doing the same thing lately! CC + godot + some game ideas I've had banging around in my head for years but daunting to dive into.

The results so far are... okay, but getting something working to validate the gameplay loop and experiment with different systems is a lot of fun!

I definitely agree that it should be shown and obvious in the UI. They do show a warning now when resuming old sessions but still could be better.

> I would also expect to see it taking exponentially longer to process a prompt. I don't believe LLMs work like that.

Try this out using a local LLM. You'll see that as the conversation grows, your prompts take longer to execute. It's not exponential but it's significant. This is in fact how all autoregressive LLMs work.

Yesterday I was playing around with Gemma4 26B A4B with a 3 bit quant and sizing it for my 16GB 9070XT:

  Total VRAM: 16GB
  Model: ~12GB
  128k context size: ~3.9GB

At least I'm pretty sure I landed on 128k... might have been 64k. Regardless, you can see the massive weight (ha) of the meager context size (at least compared to frontier models).

> As a user, I _expect_ the cost of resuming X hours/days later to be no different to resuming seconds or minutes later.

As an informed user who understands his tools, I of course expect large uncached conversations to massively eat into my token budget, since that's how all of the big LLM providers work. I also understand these providers are businesses trying to make money and they aren't going to hold every conversation in their caches indefinitely.

Definitely some problems in the current system, broad and creeping executive overreach extending back decades now.

Pretty sure stealing from stores is already illegal, not sure I understand your analogy... lots of case law / precedent there.

I do agree it was a weak case, I think SCOTUS ruled correctly.

Haven't had a chance to test 4.7 much but one of my pet peeves with 4.6 is how eager it is to jump into implementation. Though maybe the 4.7 is smarter about this now.

The system prompt is always loaded in its entirety IIUC. It's technically possible to modify it during a conversation but that would invalidate the prefill cache for the big model providers.

Is it exponential or logistic?

Nope, the original tariffs were under IEEPA, then Supreme Court ruled they didn't have authority to use IEEPA, so they had to drop those tariffs and start working on refunds. It'd only have been illegal if they kept the tariffs after the ruling.

Lot of propaganda & emotions around this straightforward chain of events.

Try it before you give up, I got plenty of AI stuff working on my 6750XT years ago.

Eh, YMMV. I was using rocm for minor AI things as far back as 2023 on an "unsupported" 6750XT [0]. Even trained some LoRAs. Mostly the issues were how many libs were cuda only.

[0] https://news.ycombinator.com/item?id=43207015

I've been using CC a decent amount the past few weeks and have never seen this malware stanza...?

Self-aware meta commentary on HNers breaking HN guidelines to cite HN guidelines.

They're complementary but also have significant overlap. Hence all the confusion and strong opinions.

> How about we standardize this idea? We can have an endpoint to tell the agents where to find this text file and API

Good news! It's already standardized and agents already know where to find it!

https://code.claude.com/docs/en/skills

Absolutely. You can get "locked in" to certain patterns / phrases just via muscle memory and familiarity. Need to balance that with a little improv to find new patterns phrases you like, and then can train those in via muscle memory.

Preach. I have 4 monitors and one is a vertical 1440x2560. Massive productivity boost - terminals running claude code, reading docs, IDE panes, anything with lots of scrolling. Highly recommend it!

Wow that's bad. Unsure if this is an outlier or typical for YC companies.

Pretty sure this is illegal now across the board.

No, missing a MMU.

> What you are offering is NOT a free tool -- it is a demo, for a tool for which you are charging $12/month

Commonly known as a "free trial"

I've found memories and state to be a mixed bag with LLMs. To the point I don't bother with long term memories - usually only short or medium-term session logs or task-focused docs.

Ooof, you're way behind - see sibling comments. Come join us in the promised land, brother!

Home Assistant integration a la Alexa would be awesome