When I saw 2 vCPU and 4GB, and then below that "4 GB persistent disk" I was like ooohhh it's the disk, not the memory. But it IS actually 4GB memory + 4GB disk.
When you say Jev is optimized for it, I get that there’s no need for unnecessary decodes in an autoregressive fashion.
But both LLMs and Jev-like models would need to prefill, the only optimization Jev does differently is the decode which can be emulated by reading off logprobs.
We don’t know the param size of Jev, to determine the most comparable model, but if I had to guess it’s sub-100B.
I also noticed this personally. Consider when an agent says
"Should I start with the spec, or go straight to building it?
Unless you're working on a super crucial piece of engineering, you can probably get away with going straight to building it. Even if something doesn't go the way you intended, I find that it's usually faster to correct the agent later, once the initial implementation is in place. It's more of an iterative approach to building and I feel that it is less cognitively demanding.
No doubt it's anything but perfect, but I can give anecdata that my M1 just slept overnight while not plugged in, and I had 99% charge just now when I opened it up to check HN.
In the case of DeepSeek-V4-Flash-0731 on deepinfra.com there was little difference when both used high reasoning. In the case of that same model on together.ai there was a substantial difference between the two (high reasoning for both again). When using together.ai with Kilo Code the LLM was having a lot of trouble making successful edits. In some cases that meant a lot tries at using the tools and in others it worked around by running scripts. Meanwhile it used the tools from my harness just fine. I've specifically tried to make my tools easy for all of the open weight LLMs to use correctly. That was inspired by getting some errors from Kilo Code at the beginning of the year telling me that the model was having trouble and I should use a smarter model.
Lots of criticism here but I think this is extremely promising. When this tech is extended to videos and perhaps even other forms of media, I think it has the potential of stopping all slop!
All slop? I'm sure that some "Shrimp Jesus" or "Talking Strawberry" was never considered to be authentic by anybody. There is a lot of useless AI generated content of which everybody knows it's AI generated littering the web. Having it marked as AI will not stop that.
As if it were that easy? Being that rich and influental means someone out there will attempt (and succeed) to game you.
reply