ilia-a

Trying to address the long-standing issue of un-safe deserialization by introducing built-in hmac into PHP serialization process and at the same time improving general performance (now better or equal to igbinary)

Unfortunately at the moment the model is very quickly burning through plans, single session with 3-4 subagents, none using Max or Xhigh, mostly medium + some High can burn through 5h limit within 20-40 minutes of usage.

First of all there is a way to made agents spot truncation by being aware of RTK compression and having bypass option (I use RTK_DISABLE=1) as a way of restoring original full text.

Works fine, yeah it only compresses command output so only input tokens are affected in terms of "compression".

PHP

I did fill the form our a while back (it was around for a few months now) without any response. I guess must be really big OSS project for maintainer to qualify.

Seems odd limit, especially since it highly dependant on Token provider used, with Opus this is not much and could easily be burnt in a week or less, but with something like deepseek the 1500 can literarily be an annual budget.

That being said, I do have to wonder why someone as bug as say Uber, simply not rollout OSS model in the cloud for their team, I'd imagine that would be cheapest & most flexible option, while also keeping all the data shared with LLM private.

I mean they are comparing themselves to Haiku of all things, geez that's not a good start...

Because of the memory costs lately, I doubt this will be much cheaper. Also this is quite a bit slower than even 4070 let alone *90 Nvidia variants albeit with much lower memory.

Looks like just rebranded DGX in laptop form, the biggest miss is the weak memory speed, 1/2 of the M5 laptop memory speed, and 1/3 of the M3 ultra that is now years old...

Yeah, it is a massive jump in price, hardly a "Flash" model anymore... I wonder if they'll release flash lite or something with a bit more affordable price point.