Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This is more exciting than k3, IMO. Dsv4 models are extremely cheap to serve. Improving their capabilities has lots of downstream effects, as it becomes "good enough" for more and more tasks.

DS was serving the pro version at extremely low prices for a long time, and they've had integrations with opencode & other providers, so they likely gathered a lot of data from real developers doing real tasks (on openrouter they were labeled as such). Now they can use those live scenarios to further post-train their models and improve them further.

Can't wait to see if distilling k3 into dsv4 brings additional improvements. Anyway, having fast cheap models getting better is great for the community. Especially since these don't "go away" on a provider's whim. Whatever capabilities they get, can be used "forever" going forward. And, at least flash can be ran "at home" with <10k in hardware, which isn't really possible / feasible with glm/k3 larger models.

 help



  > Whatever capabilities they get, can be used "forever" going forward.
"Forever" gets the scare quotes because it is implied only up until the Butlerian Jihad?

I upvoted you just for the Dune reference.

I feel like once they make a mega successful movie of a book we can stop doing the secret handshake.

OP refers to details from the book that are not in either film.

“Thou shalt not make a machine in the likeness of a human mind”…


The only movie I've ever enjoyed after reading the book was Jurassic Park. I've never seen the Dune movies or most other book-based movies, but honestly The Martian and Project Hail Mary were tempting. I'll probably surrender to one of those two someday.

The most recent Dune movies pull off the incredible feat of being a better telling of the tale than the book, but only if you already know the tale from the books. They’re not beginner-friendly, but they’re extremely respectful to the book reader, while still make conscious (and in my opinion, positive) changes to the story.

Thank you. Maybe I'll read Dune again (or listen to an audiobook) then watch.

That sounds exactly like the type of experience I'd enjoy.


I grew up on Dune lore and the first movie is one of my favorites of all time. Highly recommended for any Dune fan. I didn't care for the second one.

Especially when the movie is over forty years old.

I don't think 1984 dune was successful culturally or financially. Definitely a different beast from the recent attempt, which has been wildly popular.

Totally! This with DwarfStar delivers usable local AI (I hope!)

Usable local AI has been here for a while, esp on say a 5090.

You can’t treat Qwen3.6 like its fable, but if you prompt precisely and specifically it’s a great executor.

I actually found it refreshing to use more of my brain for once, and actually have to think deeper about what I’m trying to do, and how to build it.



What are your goalposts? Depending on your requirements, there have been many moments of usable local AI. More recent ones were gpt-oss 120b and Qwen 3.6 27b.

Starting to wonder if the free big pickle model on opencode has been DSV4F0731 for the past few months. It’s been incredibly fast and good.

At least in the past it was GLM-4.6. IDK if it is ever changed.

https://github.com/anomalyco/opencode/issues/4276


I’ve heard that too, but since it’s a stealth model, I suspect they change it to whatever preview model provider that offers a free endpoint. In the past few months, I’ve gotten 4xx errors identifying the provider as DeepSeek.

distilling K3 into DS4 flash will likely only be a good idea for specialists. the difference in model capacity is otherwise too large.

we used to distill GLM 5.2 into Qwen 27B specialists with great success.

developing the workflows is tricky though. we had the advantage of a straightforward mapping function in mind: English -> SQL which was relatively easy to pipeline training for.


Why would you do this?

a specialist model sufficiently post-trained can outperform a frontier model while being dirt cheap.

what we did was distill GLM 5.2 into a 27B model on SQL and then post-train it with RL afterward. the result outperformed even Fable on that one task. the distillation step is just good sense in this workflow, to bootstrap a smaller model to the utmost you can before actually doing RL.


Do you have any reference guides on how to perform this distillation? I have a use case for this and hardware to run it on, but have never seen how it's done.

nothing publicly available that I'm aware of.

hope that deepseek become better

Have you tried the one that was just released?



Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: