Hacker Newsnew | past | comments | ask | show | jobs | submit | replete's commentslogin

Its exciting to see, but look at the die size for only an 8b model


Run server with ollama, use Continue extension configured for ollama


I'd stay away from ollana, just use llama.cpp; it is more up date, better performing and more flexible.


But you can't just switch between installed models like in ollama, can you?



It's a practical approach, I used vagrant many years ago mostly successfully. I also explored the docker-in-docker situation recently while working on my own agentic devcontainer[0]- the tradeoffs are quite serious if you are building a secure sandbox! Data exfil is what worries me most, so I spent quite some time figuring out a decent self-contained interactive firewall. From a DX perspective, devcontainer-integrated IDEs are quite a convenient workflow, though docker has its frustrating behaviours

[0]: https://github.com/replete/agentic-devcontainer


An easy switch for an individual, for an organization the ecosystem is typically too much to give up. Hopefully this changes in a few years.


Organizations are composed of individuals, if your organization is tied to a specific IDE to be able to function, that seems like a huge problem.


Ecosystem.


If your organization's "ecosystem" is tied to a specific editor, your organization has major issues.


Do some companies/orgs mandate which editor to use? Is this usually big companies? Which countries?


They are focusing on an Azure migration for then next 2 years...


Codenames are good identifiers when lots of projects exist. There are more software projects now than EVER.

I welcome Splorg, Chizzel, Rapunzel, Brap, Titoid, and Chungus - I don't care as long as it's good.


A whitelist in package.json is only a partial assist


Because they use emdashes?


Because it's obvious.


Is this a joke?

Kids made this in the G-Mod sandbox for Half-Life 2.

If it feels dystopian, it's because HL2 is set in a dystopian world.

Honestly, it might literally be the toilet aspect that made this viral with 7 year olds


I don't think you're wrong, but also don't think that the article's wildly wrong either.


Resolution isn't the point, the semantic confusion is what generates clicks.


I think the M5 Max will be more like 614GB/s, unless they somehow have exceeded DDR5x-9600 or added more than 32 memory controllers


DDR5-9600 is 153GB/s from a single channel, Max has 4 channels… these are all theoretical values of course - real world none of these, even the graphics card will get that near to those… so not sure what you’re saying.


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

Search: