DigitalOcean is bad not just through OR... Direct accesss through DO gave me such a miniscule context and maximum tokens, it was pretty much unusable. So, accessing it through OpenRouter gave bigger context length, but model behaves like it was seriously damaged - tool calling was producing paths that had / replaced with some other symbols, files were not found etc., and this is just for things that errored out, I have no idea how bad reasoning was. I had to blacklist DO completely on OpenRouter.

    > Granted, I was trying to have it solve a hard problem that 5.6 Sol couldn't solve, but still.

Did it solve it?

The suspense is killing me!

It is still working on it!

And must not be disturbed!

Wait, you have noticed it too? I burned through weekly limit in 25 hours after last reset on Thursday, without changing what I was doing before that.

Interestingly enough, previous subscription to Plus gave me about 2-3 days of coding. So I switched to Pro Light now, gave me about the same amount for a week or so (G-d bless these quota resets of theirs!). Now, with the last reset I only gathered 25 hours before I hit my weekly limit. Now I am buying credits, I switched to Luna to execute very narrow sets of patches, and offload things to "free" Spark model, and it is blowing through tokens less actively, but noticeable quickly too. I am not sure if it is something with how tokens are counted, or how they are counted depending where you are in a subscription cycle. Could it be that one person's "limit" is not the same as another, trying to push you to buy token credits?

To end up on your phone, VOIP have to terminate somewhere in local phone network. These networks need to stop accepting calls from shady places, or every call like this that they cannot charge back to original (or intermediary network) becomes a loss to them. And telecom companies do not like losses...

Somehow reminds me of JetBrain's new Air [1] platform (still in beta I think) - make multiple agents working, monitoring, basic code and diff viewing.

[1] https://air.dev/

Well, technically it is their hardware now...

And their team, if they treat them well.

Remove permission to execute JavaScript (in your browser) on sites that use html-load.com and error-report.com in addition to blocking these two in DNS. Solved it for me, don't know for how long.

I'm thinking about using this and have a few questions:

    1. How are hardlinks and duplicate files (same content, different paths) handled?
    2. Does deduplication work on a block/chunk level for partially matching files, or does it only look at whole files?
    3. Is there any specific integration or handling for Copy-on-Write (CoW)?

Thank you!

Hardlinks work as you’d expect: multiple paths point to the same inode and data, so a write through one path is visible through the others. Two separate files with the same contents are stored separately.

There’s no deduplication, either whole-file or block-level. That’s intentional, mostly because of the impact it would have on locality.

If by CoW you mean reflinks, those aren’t currently planned either. They avoid the content matching part of deduplication, but still require sharing extents between files and come with similar locality and complexity tradeoffs. Internally ZeroFS is copy-on-write, with immutable segments and checkpoints, but that isn’t exposed as reflinks.

Thank you for the quick answer.

Which provider do you use, if you don't mind sharing? I tried Digital Ocean (looking for Zero retention and no training), but their context limits are rather small for DeepSeek inference.

I used some others but eventually decided to use Deepseek directly. I don't mind giving China my data, I'm more concerned about the American government imprisoning me than a foreign nation an ocean away.

Still fun to watch models trying their best :). I think "money spent" metric in this test is becoming the most interesting to watch. It is like looking at RT cost of "Hello, world!"...