Of course, it would be far better to just not have the memory leak...but in the mean time, macOS memory compression is making the leak nearly a none issue.
Do not buy RAM, compress (curiousbits.nl)
varenc 5 hours ago
trollbridge 4 hours ago
Sharlin 3 hours ago
bayindirh 3 hours ago
Fun days.
Now, we find out when the SSD wears out, in the worst case.
Lerc 17 minutes ago
dotancohen 2 hours ago
hobo123 an hour ago
davidkwast 4 hours ago
01HNNWZ0MV43FF 3 hours ago
Thus solving memory leaks forever...
bayindirh 3 hours ago
kccqzy 13 minutes ago
cpeterso 3 hours ago
kevincox 2 hours ago
Sure, they will get restarted, but not great.
cxr 2 hours ago
going through extra effort to accommodate/enable buggy, poorly behaving programs, and meanwhile—
punishing those (and the programmers of those) that are well-behaved and functioning exactly as intended but that now have to contend with evading some heuristic of badness.
dotancohen 2 hours ago
TheSoftwareGuy 2 hours ago
InsidiousWombat 9 minutes ago
... but reading say libraries directly and only putting in RAM the parts often used, now we're talking
fl0ki 18 minutes ago
Lio 10 hours ago
Since the models are effectively highly compressed data already my assumption is that RAM compression won't help improve that much (I'm assuming unified memory here as per the Framework Desktop). A bit like trying to compress a JPEG in a zip file.
Could anyone comment on that?
rft 10 hours ago
kalcode 10 hours ago
But all seem to have massive issue boards. What I did see was oMlx release clustering so you can string 3 macs and have 3 times the memory or use one for prefilling accelerating.
Which I thought was neat and what I'll likely poke at instead of getting more ran.
0123456789ABCDE 9 hours ago
kalcode 5 hours ago
So my mind on that.
matja 8 hours ago
Interesting comparison, I wonder if there are loss-less compression schemes that are "weight-aware", about the statistical properties of model weights.
Back to that line - because (Huffman coded) JPEG images can be lossless recoded using arithmetic coding, which makes them smaller than if they were compressed with the best general-purpose compression algorithms/tools (e.g. `jpegtran -arithmetic < input.jpg > output.jpg` beats `xz -9 < input.jpg > output.jpg`). AC takes advantage of the properties of the image coding better than general-purpose compression tools on the output of the Huffman coding.
rft 7 hours ago
strobe 5 hours ago
throw-qqqqq 5 hours ago
Most quantized GGUFs have a very high Shannon entropy, so are almost perfectly compressed already. A Q4_K_M quantization is often around +7.90 bits/byte and so is virtually incompressible.
Unquantized models (F16) are often around +7bits/byte for comparison.
https://en.wikipedia.org/wiki/Shannon%27s_source_coding_theo...
vlovich123 5 hours ago
One way to think about it is that if the weights were compressible, it’s a clear signal that better training could make the model smarter as there’s redundant information in the structure that could be used to store other meaning.
The other challenge with compressing weights of course is that you have to decompress them before you do math with them. Dequantizing weights is significantly cheaper and supports random access too (not that that matters too much but it helps in terms of leveraging SIMD techniques to do it in an embarrassingly parallel way)
Powdering7082 4 hours ago
Consider the induction heads, their W_O are generally expected to be the identity matrix and it is useful to have many of them at many layers (e.g. if you only had one induction head in the whole network you'd only be able to perform one induction circuit per token emitted)
That being said the attention heads are a small portion of the total network size
hobo123 an hour ago
I'm not too familiar with LLM memory access patterns, but since the model files are stored on disk, you wouldn't ever keep them in memory anyway (except caching) and just reread them on demand.
RossBencina 31 minutes ago
Dense architectures hit every byte of the model for every token. It's the ultimate memory bandwidth nightmare.
MoE models use subsections for each token and are more amenable to keeping parts in secondary storage, and indeed there are some local llm engines that do this.
jakozaur 10 hours ago
Yes, macOS has had it enabled by default since OS X Mavericks 10.9, released in October 2013. I’m surprised some desktop Linux distributions still require manual configuration to enable it.
For browser tabs, though, I think a better solution is automatic tab closing, using something like the Tab Wrangler extension for Chrome. Collecting tabs comes naturally; we’re much worse at closing them.
jve 9 hours ago
evenfrost 9 hours ago
jamesgeck0 5 hours ago
kasabali 4 hours ago
aa-jv 9 hours ago
Its a user choice and should be a user-facing knob because there are times it is inappropriate for use, such as in safety-critical circumstances, certain realtime realities, and so on. It should probably be on by default for desktop Linux, but off by default for every other application a distribution may be targeting ...
xnx 6 hours ago
benjiro29 5 hours ago
Do Shift+Esc and watch how large the GPU memory usage is. Often 2/3 to 50% of your actual tab usage.
Ironically, its not the closing of tabs that bothers me, but the fact that tab order changes, what slows my productivity/muscle memory down. And its the same reason for a lot of people why they keep tabs open.
I never understood why we do not have a basic snapshot / sleep that offloads a tabs memory on disk (and dumps the GPU memory), aka a ghost tab. And fast loads when you view the tab. Sure, there may be a small delay but it can save so much memory, while people can keep their tabs open/organized.
Some extensions try to mimic that but its often a crude solution because they do not have deeper access to the webview2 engine.
hobo123 an hour ago
In Firefox there's about:unloads where you can suspend tabs in I believe LRU order. I'd really like to see this as a default option (like: only keep N tabs active). Maybe there's an extension for that.
hedgehog 4 hours ago
flowerlad 3 hours ago
[1] https://en.wikipedia.org/wiki/SoftRAM
[2] https://devblogs.microsoft.com/oldnewthing/20211111-00/?p=10...
cogman10 6 hours ago
https://chrisdown.name/2026/03/24/zswap-vs-zram-when-to-use-...
ciupicri 5 hours ago
mrob 4 hours ago
dsp_person 4 hours ago
01HNNWZ0MV43FF 3 hours ago
quotemstr an hour ago
mrob an hour ago
quotemstr 27 minutes ago
All you're doing by running without swap is pessimizing yourself.
InsidiousWombat 7 minutes ago
teravor 3 hours ago
if you don't want to swap to disk you should not use zswap.
zram is ideal in that scenario, though it may complicate OOM killer considerations (you may want a userspace one).
jmakov 2 hours ago
If my understanding is correct, both ZRAM and ZSWAP write uncompressed pages to disk. And both don't pack the pages together meaning even if we'd write compressed pages (let's say they compress 5:1), we'd still write the original size to disk. This is crazy.
quotemstr an hour ago
zamalek an hour ago
ZRAM added compressed_writeback in 7.0, it needs to be explicitly enabled.
jmakov an hour ago
noodlesUK an hour ago
zamalek an hour ago
I recall reading an article around the time the Windows team added it (back when I still used the OS). Their compression algorithm was typically faster than the latency to the prevailing disk type at the time (SATA SSD, most likely) - at the time it had zero downsides. I'm not sure if that still holds true.
On Linux it generally doesn't: off-the-shelf compression is used, which do not outpace storage access latency.
sammatilda 10 hours ago
I am hoping to switch my dad from ubuntu to fedora in a month or so because he has 16 GB ram and this seems like something most people dont know.
nico 10 hours ago
Lio 10 hours ago
I really like Fedora in that it seems very up to date package wise but I found it requires a reboot every time I update using the UI updater.
It's better using the dnf command but it still felt a bit lumpy compared to apt.
mritzmann 10 hours ago
nosrepa 2 hours ago
speed_spread 10 hours ago
amelius 9 hours ago
jrmg 6 hours ago
swiftcoder 5 hours ago
mkesper 5 hours ago
anthk 4 hours ago
kasabali 4 hours ago
quotemstr 31 minutes ago
hofiflo 4 hours ago
adgjlsfhk1 3 hours ago
hofiflo an hour ago
hobo123 an hour ago
InsidiousWombat 5 minutes ago
Juvination 5 hours ago
2001zhaozhao an hour ago
myrandomcomment 5 hours ago
certyfreak 10 hours ago
nukanse 9 hours ago
bullish-bearr 39 minutes ago
Lerc 6 hours ago
I find that the ratio of browserish memory use to that of other programs is enough to render the latter insignificant most of the time.
anthk 4 hours ago
roosgit 12 hours ago
kachnuv_ocasek 10 hours ago
ChocolateGod 4 hours ago
anthk 4 hours ago
mrob 3 hours ago
aktenlage 9 hours ago
Isn't that wasteful? The CPU's caches may help a bit, but I would guess first a huge block is going through the decompression, so only the last bits are still in the cache, when the actual usage starts.
altcognito 5 hours ago
If it avoids hitting the disk I get it, but to some of us, that just means it's time to buy more RAM
rustyhancock 10 hours ago
What's interesting to note is that I've occasionally wondered why isn't the swap partition compressed.
If you have it on an NVMe drive it probably is as is all data! The controller usually transparently does it as it writes to/from NAND.
So the kernel sees uncompressed fixed block pages and the speed boosts happen with data arriving faster when read or written to flash.
close04 6 hours ago
I started using zswap recently but didn't have the time yet to measure the benefit. Does it benefit machines with low amounts of RAM and wimpy CPU (like a RPi3)?
> The controller usually transparently does it as it writes to/from NAND.
In the consumer SSD space the last controllers I know did on-the-fly compression were SandForce [1], probably 15 years ago, when minimizing writes and saving the little capacity available was imperative. I don't think any other desktop SSDs do this now. A bit more common in enterprise models like ScaleFlux CSD.
mkesper 5 hours ago
deltoidmaximus 3 hours ago
pyaamb 5 hours ago
NegativeLatency 5 hours ago
Tried porting it to swift, was hoping it would make mobile local models more useful for a project I was working on but it was hurting the accuracy too much. Can be very task dependent on how much you can compress. https://github.com/nburns/llmlingua-swift
My specific use case was using a LLM on segments of labeled data from a different faster model as a final pass.
grandlimelabs 3 hours ago
IronWolve 4 hours ago
1970-01-01 3 hours ago