I was a big fan of deno, so much that I migrated all my cf worker to demo, cause you can run the same engine in there cloud and local/selfhost - perfect no vender lockin(like with cloud flare workers), it's so sad to see it gets shelved.
I really really love deepseek v4(and 4.1), but for everything I thrown at it, it felt like gpt 5.6 + 6 luna or terra do it faster, in way less tokens and in a way I like more. So even tho the price is (very) cheap, i found myself using it less just cause I don't want to wait on something I have to itteratee on with the model.
TeamSpeak, git backup(I use gittea) to backup public repos, backups in general, my website/blog, game servers, ssh(obv) for local app port forward, litellm as ai bridge to monitor usage, some of my own apps: (notification broker to send messages to my devices using a simple http push), everything behind traefik (with let's e encrypt)
The very first page[1] of the VSCode remote documentation makes it perfectly clear that VSCode remoting is more akin to running an Emacs server on the remote than TRAMP.
a hile ago (when big providers still provided logprobs) i created a VS Code highlighter that visualizes unsure tokens.
Since most chat models want to answer with a human-readable message i think their logprobs are not as meaningful. It would be interesting to see if one choice is like "correct" and if the model wants to choose it more often, cause it might not answer the question but to prose to the user.
I saw a lot of ppl think about what jev could use under the hood and could someone explain why this can't just be an embedding model where we just embed all the input + decisions and give back the cosine (or whatever) similarities?
if you compare an embedding model to something like Jev which asks 100 questions and use the answers as the embedding you will be able to get move mileage out of the latter. especially because you don't need to train any classifiers for your task, you can work directly on the answers.
that said, I don't understand the hype. I have been doing what Jev does for 2 years now by just forcing json tokens onto an LLM. you can even get the LLM to think. and you can ensemble multiple LLMs.
I suppose the appeal of Jev is how cheap and fast it is, but then it's entirely unsuitable for anything but the most cursory extraction. using it to play games seems like a waste of time especially when most of those games will be played better by an algorithm written by an LLM (just give it the state and ask it to write a bot).
> I suppose the appeal of Jev is how cheap and fast it is
Ok, so you do understand the hype. I mean cost & speed are reaaaaally big issues for normal LLMs. If you can fill a specific use case way cheaper and way faster, that’s a fantastic development. We need more niche tooling that’s more efficient and better for niche use cases. Not everything has to be general purpose.
Not clear to me that playing games is the point of jev. This post is just a random fun experiment someone did because they wanted to.
But I can ask 100 causally masked questions against common prefix, and get 100 answers, all in a single PP pass using any existing "classical" attention transformer model? Like, I had the impression that is what everyone was doing for classification already?
Is the difference "we did RL to tune logit distribution"? Because I really do not see anything new there. What is the difference?
Embeddings just convert the tokens to a vector that represents the text in an abstract semantic space. JEV goes a step further and actually processes the instructions/meaning of those embeddings to produce output, just not the usual series-of-tokens output we expect from an LLM.
reply