News for hackers. Stuff that matters.
Re:Claude Opus 4.7 by Tiberium on Thursday April 16, 2026 @10:47AM (#47793859)
I hope people realize that tools like caveman are mostly joke/prank projects - almost the entirety of the context spent is in file reads (for input) and reasoning (in output), you will barely save even 1% with such a tool, and might actually confuse the model more or have it reason for more tokens because it'll have to formulate its respone in the way that satisfies the requirements.
Re:Claude Opus 4.7 by embedding-shape on Thursday April 16, 2026 @11:10AM (#47794345)
> I hope people realize that tools like caveman are mostly joke/prank projects

This seems to be a common thread in the LLM ecosystem; someone starts a project for shits and giggles, makes it public, most people get the joke, others think it's serious, author eventually tries to turn the joke project into a VC-funded business, some people are standing watching with the jaws open, the world moves on.

Re:Claude Opus 4.7 by simonw on Thursday April 16, 2026 @11:33AM (#47794813)
I was convinced https://github.com/memvid/memvid was a joke until it turned out it wasn't.
Re:Claude Opus 4.7 by embedding-shape on Thursday April 16, 2026 @11:36AM (#47794857)
To be fair, most of us looked at GPT1 and GPT2 as fun and unserious jokes, until it started putting together sentences that actually read like real text, I remember laughing with a group of friends about some early generated texts. Little did we know.
Re:Claude Opus 4.7 by Alifatisk on Thursday April 16, 2026 @11:52AM (#47795172)
Are there any public records I can see from GPT1 and GPT2 output and how it was marketed?
Re:Claude Opus 4.7 by walthamstow on Thursday April 16, 2026 @11:56AM (#47795254)
I don't think it was marketed as such, they were research projects. GPT-3 was the first to be sold via API
Re:Claude Opus 4.7 by embedding-shape on Thursday April 16, 2026 @12:22PM (#47795717)
HN submissions have a bunch of examples in them, but worth remembering they were released as "Look at this somewhat cool and potentially useful stuff" rather than what we see today, LLMs marketed as tools.

https://news.ycombinator.com/item?id=21454273 / https://news.ycombinator.com/item?id=19830042 - OpenAI Releases Largest GPT-2 Text Generation Model

HN search for GPT between 2018-2020, lots of results, lots of discussions: https://hn.algolia.com/?dateEnd=1577836800&dateRange=custom&...

Re:Claude Opus 4.7 by Den_VR on Thursday April 16, 2026 @11:36PM (#47802237)
I still think of The Unreasonable Effectiveness of Recurrent Neural Networks and related writings.

http://karpathy.github.io/2015/05/21/rnn-effectiveness/

Re:Claude Opus 4.7 by embedding-shape on Friday April 17, 2026 @07:21AM (#47804702)
Fun to revisit no doubt, the comments make it even better.

> SuckCocker 7 years ago - "in short: SKYNET is not far away. Be proud to be a part of it!"

Re:Claude Opus 4.7 by dalemhurley on Friday April 17, 2026 @12:14AM (#47802397)
Wild how many people were predicting the AI slop, but was dismissing it as unlikely beyond some trolls.
Re:Claude Opus 4.7 by maplethorpe on Thursday April 16, 2026 @12:27PM (#47795808)
From a 2019 news article:

> New AI fake text generator may be too dangerous to release, say creators

> The Elon Musk-backed nonprofit company OpenAI declines to release research publicly for fear of misuse.

> OpenAI, an nonprofit research company backed by Elon Musk, Reid Hoffman, Sam Altman, and others, says its new AI model, called GPT2 is so good and the risk of malicious use so high that it is breaking from its normal practice of releasing the full research to the public in order to allow more time to discuss the ramifications of the technological breakthrough.

https://www.theguardian.com/technology/2019/feb/14/elon-musk...

Re:Claude Opus 4.7 by ethbr1 on Thursday April 16, 2026 @12:38PM (#47795966)
Aka 'We cared about misuse right up until it became apparent that was profit to be had'

OpenAI sure speed ran the Google and Facebook 'Don't be evil' -> 'Optimize money' transition.

Re:Claude Opus 4.7 by sfn42 on Thursday April 16, 2026 @01:39PM (#47796863)
Or - making sensational statements gets attention. A dangerous tool is necessarily a powerful tool, so that statement is pretty much exactly what you'd say if you wanted to generate hype, make people excited and curious about your mysterious product that you won't let them use.
Re:Claude Opus 4.7 by eric_h on Thursday April 16, 2026 @02:24PM (#47797440)
Much like what Anthropic very recently did re: Mythos
Re:Claude Opus 4.7 by xpe on Thursday April 16, 2026 @05:35PM (#47799798)
Think about all the possible explanations carefully. Weight them based on the best information you have.

(I think the most likely explanation for Mythos is that it's asymmetrically a very big deal. Come to your own conclusions, but don't simply fall back on the "oh this fits the hype pattern" thought terminating cliché.)

Also be aware of what you want to see. If you want the world to fit your narrative, you're more likely construct explanations for that. (In my friend group at least, I feel like most fall prey to this, at least some of the time, including myself. These people are successful and intelligent by most measures.)

Then make a plan to become more disciplined about thinking clearly and probabilistically. Make it a system, not just something you do sometimes. I recommend the book "the Scout Mindset".

Concretely, if one hasn't spent a couple of quality hours really studying AI safety I think one is probably missing out. Dan Hendrycks has a great book.

Re:Claude Opus 4.7 by wat10000 on Thursday April 16, 2026 @12:29PM (#47795844)
You can run GPT2! Here's the medium model: https://huggingface.co/openai-community/gpt2-medium

I will now have it continue this comment:

I've been running gps for a long time, and I always liked that there was something in my pocket (and not just me). One day when driving to work on the highway with no GPS app installed, I noticed one of the drivers had gone out after 5 hours without looking. He never came back! What's up with this? So i thought it would be cool if a community can create an open source GPT2 application which will allow you not only to get around using your smartphone but also track how long you've been driving and use that data in the future for improving yourself...and I think everyone is pretty interested.

[Updated on July 20] I'll have this running from here, along with a few other features such as: - an update of my Google Maps app to take advantage it's GPS capabilities (it does not yet support driving directions) - GPT2 integration into your favorite web browser so you can access data straight from the dashboard without leaving any site! Here is what I got working.

[Updated on July 20]

Re:Claude Opus 4.7 by fancyfredbot on Thursday April 16, 2026 @04:51PM (#47799336)
Wow that is terrible. In my memory GPT 2 was more interesting than that. I remember thinking it could pass a Turing test but that output is barely better than a Markov chain.

I guess I was using the large model?

Re:Claude Opus 4.7 by sillysaurusx on Thursday April 16, 2026 @06:27PM (#47800334)
There’s an art to GPT sampling. You have to use temperature 0.7. People never believe it makes such a massive difference, but it does.
Re:Claude Opus 4.7 by wat10000 on Thursday April 16, 2026 @05:32PM (#47799765)
Probably a much better prompt, too. I just literally pasted in the top part of my comment and let fly to see what would happen.
Re:Claude Opus 4.7 by daveguy on Thursday April 16, 2026 @05:23PM (#47799673)
Here is the XL model. 20x the size of the medium model. Still just 2B parameters, but on the bright side it was trained pre-wordslop.

https://huggingface.co/openai-community/gpt2-xl

Re:Claude Opus 4.7 by mlsu on Thursday April 16, 2026 @12:55PM (#47796223)
I was first made aware of GPT2 from reading Gwern -- "huh, that sounds interesting" -- but really didn't start really reading model output until I saw this subreddit:

https://www.reddit.com/r/SubSimulatorGPT2/

There is a companion Reddit, where real people discuss what the bots are posting:

https://www.reddit.com/r/SubSimulatorGPT2Meta/

You can dig around at some of the older posts in there.

Re:Claude Opus 4.7 by PufPufPuf on Thursday April 16, 2026 @03:19PM (#47798181)
I used GPT-2 (fine-tuned) to generate Peppa Pig cartoons, it was cutely incoherent https://youtu.be/B21EJQjWUeQ
Re:Claude Opus 4.7 by Bombthecat on Thursday April 16, 2026 @12:13PM (#47795573)
And now gpt is laughing,while it replaces coders lol
Re:Claude Opus 4.7 by DiffTheEnder on Thursday April 16, 2026 @12:17PM (#47795646)
Is this... just one file acting as memory?
Re:Claude Opus 4.7 by paulddraper on Thursday April 16, 2026 @07:41PM (#47800912)
One video file
Re:Claude Opus 4.7 by MarcelOlsz on Thursday April 16, 2026 @11:44AM (#47795027)
Why? Doesn't have jokey copy. Any thoughts on claude-mem[0] + context-mode[1]?

[0] https://github.com/thedotmack/claude-mem

[1] https://github.com/mksglu/context-mode

Re:Claude Opus 4.7 by simonw on Thursday April 16, 2026 @12:00PM (#47795342)
The big idea with Memvid was to store embedding vector data as frames in a video file. That didn't seem like a serious idea to me.
Re:Claude Opus 4.7 by nico on Thursday April 16, 2026 @12:19PM (#47795669)
Very cool idea. Been playing with a similar concept: break down one image into smaller self-similar images, order them by data similarity, use them as frames for a video

You can then reconstruct the original image by doing the reverse, extracting frames from the video, then piecing them together to create the original bigger picture

Results seem to really depend on the data. Sometimes the video version is smaller than the big picture. Sometimes it’s the other way around. So you can technically compress some videos by extracting frames, composing a big picture with them and just compressing with jpeg

Re:Claude Opus 4.7 by jermaustin1 on Thursday April 16, 2026 @12:20PM (#47795691)
> embedding vector data as frames in a video file

Interesting, when I heard about it, I read the readme, and I didn't take that as literal. I assumed it was meant as we used video frames as inspiration.

I've never used it or looked deeper than that. My LLM memory "project" is essentially a `dict<"about", list<"memory">>` The key and memories are all embeddings, so vector searchable. I'm sure its naive and dumb, but it works for my tiny agents I write.

Re:Claude Opus 4.7 by niuzeta on Thursday April 16, 2026 @11:50AM (#47795127)
Just read through the readme and I was fairly sure this was a well-written satire through "Smart Frames".

Honestly part of me still thinks this is a satire project but who knows.

Re:Claude Opus 4.7 by pred_ on Friday April 17, 2026 @09:39AM (#47805843)
Time for https://itsid.cloud/index2.html to be acquired by one of the big players, I guess.
Re:Claude Opus 4.7 by imiric on Thursday April 16, 2026 @11:46AM (#47795047)
A major reason for that is because there's no way to objectively evaluate the performance of LLMs. So the meme projects are equally as valid as the serious ones, since the merits of both are based entirely on anecdata.

It also doesn't help that projects and practices are promoted and adopted based on influencer clout. Karpathy's takes will drown out ones from "lesser" personas, whether they have any value or not.

Re:Claude Opus 4.7 by combobyte on Thursday April 16, 2026 @01:20PM (#47796584)
> most people get the joke

I hope you're right, but from my own personal experience I think you're being way too generous.

Re:Claude Opus 4.7 by dakolli on Thursday April 16, 2026 @01:39PM (#47796855)
Its the same as cyrpto/nft hype cyles, except this time one of the joke projects is going to crash the economy.
Re:Claude Opus 4.7 by msikora on Thursday April 16, 2026 @06:01PM (#47800085)
This has been a thing way before AI. Anyone remembers Yo, the single button social media app that raised $1M in 2014?
Re:Claude Opus 4.7 by ghgr on Thursday April 30, 2026 @04:43AM (#47959839)
And not only in the LLM ecosystem. Flask was originally an April Fool's joke too.

https://news.ycombinator.com/item?id=13436724

Re:Claude Opus 4.7 by stingraycharles on Thursday April 16, 2026 @11:21AM (#47794549)
While the caveman stuff is obviously not serious, there is a lot of legit research in this area.

Which means yes, you can actually influence this quite a bit. Read the paper “Compressed Chain of Thought” for example, it shows it’s really easy to make significant reductions in reasoning tokens without affecting output quality.

There is not too much research into this (about 5 papers in total), but with that it’s possible to reduce output tokens by about 60%. Given that output is an incredibly significant part of the total costs, this is important.

https://arxiv.org/abs/2412.13171

Re:Claude Opus 4.7 by altruios on Thursday April 16, 2026 @12:05PM (#47795427)
Who would suspect that the companies selling 'tokens' would (unintentionally) train their models to prefer longer answers, reaping a HIGHER ROI (the thing a publicly traded company is legally required to pursue: good thing these are all still private...)... because it's not like private companies want to make money...
Re:Claude Opus 4.7 by stingraycharles on Thursday April 16, 2026 @01:55PM (#47797093)
I don’t think this is a plausible argument, as they’re generally capacity constrained, and everyone would like shorter (= faster) responses.

I’m fairly certain that in a few more releases we’ll have models with shorter CoT chains. Whether they’ll still let us see those is another question, as it seems like Anthropic wants to start hiding their CoT, potentially because it reveals some secret sauce.

Re:Claude Opus 4.7 by Ifkaluva on Thursday April 16, 2026 @09:27PM (#47801588)
I guess mainly they don’t want you to distill on their CoT
Re:Claude Opus 4.7 by stingraycharles on Friday April 17, 2026 @06:46AM (#47804494)
Yes, which I understand, but I think they’re crippling their product for users this way.

I don’t think it’s just this, because the thinking tokens often reveal more about Anthropic’s inner workings. For example, it’s how the whole existence of Claude’s soul document was reverse engineered, it often leaks details about “system reminders” (eg long conversation reminders).

I think it’s also just very convenient for Anthropic to do this. The fact that they’re also presenting this as a “performance optimization” suggests they’re not giving the real reason they do this.

Re:Claude Opus 4.7 by gwern on Thursday April 16, 2026 @03:12PM (#47798091)
LLM APIs sell on value they deliver to the user, not the sheer number of tokens you can buy per $. The latter is roughly labor-theory-of-value levels of wrong.
Re:Claude Opus 4.7 by fancyfredbot on Thursday April 16, 2026 @05:00PM (#47799441)
Try setting up one laundry which charges by the hour and washes clothes really really slowly, and another which washes clothes at normal speed at cost plus some margin similar to your competitors.

The one which maximizes ROI will not be the one you rigged to cost more and take longer.

Re:Claude Opus 4.7 by sebastiennight on Thursday April 16, 2026 @06:50PM (#47800514)
I don't think the analogy is correct here.

Directionally, tokens are not equivalent to "time spent processing your query", but rather a measure of effort/resource expended to process your query.

So a more germane analogy would be:

What if you set up a laundry which charges you based on the amount of laundry detergent used to clean your clothes?

Sounds fair.

But then, what if the top engineers at the laundry offered an "auto-dispenser" that uses extremely advanced algorithms to apply just the right optimal amount of detergent for each wash?

Sounds like value-added for the customer.

... but now you end up with a system where the laundry management team has strong incentives to influence how liberally the auto-dispenser will "spend" to give you "best results"

Re:Claude Opus 4.7 by bombcar on Thursday April 16, 2026 @10:02PM (#47801763)
Shades of “repeat” in lather, rinse, repeat.
Re:Claude Opus 4.7 by ACCount37 on Thursday April 16, 2026 @11:25AM (#47794625)
Some labs do it internally because RLVR is very token-expensive. But it degrades CoT readability even more than normal RL pressure does.

It isn't free either - by default, models learn to offload some of their internal computation into the "filler" tokens. So reducing raw token count always cuts into reasoning capacity somewhat. Getting closer to "compute optimal" while reducing token use isn't an easy task.

Re:Claude Opus 4.7 by stingraycharles on Thursday April 16, 2026 @11:30AM (#47794739)
Yeah the readability suffers, but as long as the actual output (ie the non-CoT part) stays unaffected it’s reasonably fine.

I work on a few agentic open source tools and the interesting thing is that once I implemented these things, the overall feedback was a performance improvement rather than performance reduction, as the LLM would spend much less time on generating tokens.

I didn’t implement it fully, just a few basic things like “reduce prose while thinking, don’t repeat your thoughts” etc would already yield massive improvements.

Re:Claude Opus 4.7 by AdamN on Thursday April 16, 2026 @11:35AM (#47794840)
Yeah you could easily imagine stenography like inputs and outputs for rapid iteration loops. It's also true that in social media people already want faster-to-read snippets that drop grammar so the desire for density is already there for human authors/readers.
Re:Claude Opus 4.7 by ieie3366 on Thursday April 16, 2026 @11:17AM (#47794469)
All LLMs also effectively work by ”larping” a role. You steer it towards larping a caveman and well.. let’s just say they weren’t known for their high iq
Re:Claude Opus 4.7 by roughly on Thursday April 16, 2026 @11:28AM (#47794695)
Fun fact: Neanderthals actually had larger brains than Homo Sapiens! Modern humans are thought to have outcompeted them by working better together in larger groups, but in terms of actual individual intelligence, Neanderthals may have had us beat. Similarly, humans have been undergoing a process of self-domestication over the last couple millenia that have resulted in physiological changes that include a smaller brain size - again, our advantage over our wilder forebearers remains that we're better in larger social groups than they were and are better at shared symbolic reasoning and synchronized activity, not necessarily that our brains are more capable.

(No, none of this changes that if you make an LLM larp a caveman it's gonna act stupid, you're right about that.)

Re:Claude Opus 4.7 by adwn on Thursday April 16, 2026 @11:58AM (#47795312)
I thought we were way past the "bigger brain means more intelligence" stage of neuroscience?
Re:Claude Opus 4.7 by nomel on Thursday April 16, 2026 @12:19PM (#47795677)
All data shows there's a moderate correlation.
Re:Claude Opus 4.7 by waffletower on Thursday April 16, 2026 @12:24PM (#47795758)
Even neuronal density is simplistic, and the dimension of size alone doesn't consider that.
Re:Claude Opus 4.7 by seba_dos1 on Thursday April 16, 2026 @12:27PM (#47795807)
Bigger brain does not automatically mean more intelligence, but we have reasons to suspect that homo neanderthalensis may have been more intelligent than contemporary homo sapiens other than bigger brains.
Re:Claude Opus 4.7 by dtech on Thursday April 16, 2026 @01:12PM (#47796481)
You can't draw conclusions on individuals, but at a species level bigger brain, especially compared to body size, strongly correlates with intelligence
Re:Claude Opus 4.7 by Hikikomori on Thursday April 16, 2026 @11:25AM (#47794616)
Modern humans were also cavemen.
Re:Claude Opus 4.7 by DiogenesKynikos on Thursday April 16, 2026 @11:23AM (#47794592)
This is why ancient Chinese scholar mode (also extremely terse) is better.
Re:Claude Opus 4.7 by reacharavindh on Thursday April 16, 2026 @12:00PM (#47795345)
This specific form may be a joke, but token conscious work is becoming more and more relevant.. Look at https://github.com/AgusRdz/chop

And

https://github.com/toon-format/toon

Re:Claude Opus 4.7 by alex7o on Thursday April 16, 2026 @12:52PM (#47796178)
Also https://github.com/rtk-ai/rtk but some people see that changing how commands output stuff can confuse some models
Re:Claude Opus 4.7 by bensyverson on Thursday April 16, 2026 @11:22AM (#47794576)
Exactly. The model is exquisitely sensitive to language. The idea that you would encourage it to think like a caveman to save a few tokens is hilarious but extremely counter-productive if you care about the quality of its reasoning.
Re:Claude Opus 4.7 by andai on Thursday April 16, 2026 @06:21PM (#47800278)
Does this imply that if you train it on Gwern style output, the quality will improve?
Re:Claude Opus 4.7 by gwern on Thursday April 16, 2026 @06:57PM (#47800579)
Unfortunately, that is an oversimplification for a highly RLed/chatbot trained LLM like Claude-4.7-opus. It may have started life as a base model (where prompting it with correctly spelled prompts, or text from 'gwern', would - and did with davinci GPT-3! - improve quality), but that was eons ago. The chatbots are largely invariant to that kind of prompt trickery, and just try to do their best every time. This is why those meme tricks about tips or bribery or my-grandmother-will-die stop working.
Re:Claude Opus 4.7 by Waterluvian on Thursday April 16, 2026 @11:31AM (#47794763)
Help me understand: I get that the file reading can be a lot. But I also expand the box to see its “reasoning” and there’s a ton of natural language going on there.
Re:Claude Opus 4.7 by egorfine on Thursday April 16, 2026 @11:14AM (#47794404)
They are indeed impractical in agentic coding.

However in deep research-like products you can have a pass with LLM to compress web page text into caveman speak, thus hugely compressing tokens.

Re:Claude Opus 4.7 by claytongulick on Thursday April 16, 2026 @11:29AM (#47794710)
I don't understand how this would work without a huge loss in resolution or "cognitive" ability.

Prediction works based on the attention mechanism, and current humans don't speak like cavemen - so how could you expect a useful token chain from data that isn't trained on speech like that?

I get the concept of transformers, but this isn't doing a 1:1 transform from english to french or whatever, you're fundamentally unable to represent certain concepts effectively in caveman etc... or am I missing something?

Re:Claude Opus 4.7 by egorfine on Thursday April 16, 2026 @01:00PM (#47796294)
Good catch actually.

Okay maybe not exactly caveman dialect, but text compression using LLM is definitely possible to save on tokens in deep research.

Re:Claude Opus 4.7 by make3 on Thursday April 16, 2026 @10:51AM (#47793923)
I wonder if you can have it reason in caveman
Re:Claude Opus 4.7 by 0123456789ABCDE on Thursday April 16, 2026 @11:01AM (#47794163)
would you be surprised if this is what happens when you ask it to write like one?

folks could have just asked for _austere reasoning notes_ instead of "write like you suffer from arrested development"

Re:Claude Opus 4.7 by Sohcahtoa82 on Thursday April 16, 2026 @11:22AM (#47794557)
> "write like you suffer from arrested development"

My first thought was that this would mean that my life is being narrated by Ron Howard.

Re:Claude Opus 4.7 by micromacrofoot on Thursday April 16, 2026 @12:10PM (#47795508)
I mean we had a shoe company pivot to AI and raise their stock value by 300%, how can we even know anymore
Re:Claude Opus 4.7 by bombcar on Thursday April 16, 2026 @10:06PM (#47801779)
Lemonade and blockchain rides again!

Or was it ice tea?

Re:Claude Opus 4.7 by acedTrex on Thursday April 16, 2026 @10:50AM (#47793914)
You really think the 33k people that starred a 40 line markdown file realize that?
Re:Claude Opus 4.7 by andersa on Thursday April 16, 2026 @11:14AM (#47794405)
You mean the 33k bots that created a nearly linear stars/day graph? There's a dip in the middle, but it was very blatant at the start (and now)
Re:Claude Opus 4.7 by verdverm on Thursday April 16, 2026 @11:04AM (#47794221)
Stars are more akin to bookmarks and likes these days, as opposed to a show of support or "I use this"
Re:Claude Opus 4.7 by zbrozek on Thursday April 16, 2026 @11:10AM (#47794331)
I use them like bookmarks.
Re:Claude Opus 4.7 by LPisGood on Thursday April 16, 2026 @11:11AM (#47794354)
I use them as likes
Re:Claude Opus 4.7 by giraffe_lady on Thursday April 16, 2026 @11:12AM (#47794381)
I intentionally throw some weird ones on there just in case anyone is actually ever checking them. Gotta keep interviewers guessing.
Re:Claude Opus 4.7 by pdntspa on Thursday April 16, 2026 @11:20AM (#47794529)
The amount of cargo culting amongst AI halfwits (who seem to have a lot of overlap with influencers and crypto bros) is INSANE

I mean just look at the growth of all these "skills" that just reiterate knowledge the models already have

Re:Claude Opus 4.7 by addandsubtract on Thursday April 16, 2026 @12:45PM (#47796088)
We started out with oobabooga, so caveman is the next logical evolution on the road to AGI.
Re:Claude Opus 4.7 by causal on Thursday April 16, 2026 @12:49PM (#47796145)
Output tokens are more expensive
Re:Claude Opus 4.7 by sidrag22 on Thursday April 16, 2026 @02:02PM (#47797199)
I hesitated 100% when i saw caveman gaining steam, changing something like this absolutely changes the behaviour of the models responses, simply including like a "lmao" or something casual in any reply will change the tone entirely into a more relaxed style like ya whatever type mode.

I think a lot of people echo my same criticism, I would assume that the major LLM providers are the actual winners of that repo getting popular as well, for the same reason you stated.

> you will barely save even 1% with such a tool

For the end user, this doesnt make a huge impact, in fact it potentially hurts if it means that you are getting less serious replies from the model itself. However as with any minor change across a ton of users, this is significant savings for the providers.

I still think just keeping the model capable of easily finding what it needs without having to comb through a lot of files for no reason, is the best current method to save tokens. it takes some upfront tokens potentially if you are delegating that work to the agent to keep those navigation files up to date, but it pays dividends when future sessions your context window is smaller and only the proper portions of the project need to be loaded into that window.

Re:Claude Opus 4.7 by SEJeff on Thursday April 16, 2026 @04:14PM (#47798878)
I believe tools like graphify cut down the tokens in thinking dramatically. It makes a knowledge graph and dumps it into markdown that is honestly awesome. Then it has stubs that pretend to be some tools like grep that read from the knowledge graph first so it does less work. Easy to setup and use too. I like it.

https://graphify.net/

Re:Claude Opus 4.7 by sambellll on Thursday April 16, 2026 @04:56PM (#47799392)
Someone should make an MCP that parses every non-code file before it hits claude to turn it into caveman talk
Re:Claude Opus 4.7 by xnx on Thursday April 16, 2026 @08:40PM (#47801277)
There's a tremendous amount of superstition around LLMs. Remember when "prompt engineering" "best practices" were to say you were offering a tip or some other nonsense?