ciaranmca

Don’t think I have ever came across a queue to buy a laptop before but congrats to the framework team.

I actually really like that inverse Kickstarter idea, makes a lot of sense especially if you did some kind of enter some pain/problem, search through existing idea markets and you can throw money towards ideas that would solve whatever pain point you have. Builders would essentially just have a market of validated ideas and could submit a ‘bid’ before a set deadline and the finders would vote on which is best then the funds would resolve to whatever builder made the best product.

I would say Sent2 but there is examples of recent censorship with that as well (https://hntrbrk.com/eu-satellites/)

Never thought about the impact of comments, perhaps there is value in stripping those out of read file tools

It’s a really interesting project, worth checking out the video where he merged it with a companion computer to do computer vision tasks and inject controls straight into the flight controller https://m.youtube.com/watch?v=uaY2G5Kbj_g

True, certainly something you could do with Tailscale+tmux but if this makes that a lot easier then it could end up being popular, increasingly the people who are pushing the most money into the likes of Anthropic are not rhe people focusing too much on the amount of money being put towards achieving their goal, rather the progress(tbh more realistic to say perceived progress) towards that goal in the shortest amount of time.

Obligatory https://news.ycombinator.com/item?id=9224

Fair enough but the examples shown could surely just display some pre-cooked examples to give a demo of how rhe product works with no real cost to you or barrier to potential users.

Confused me at first as when I saw mention of local + the single file thing in the GitHub I assumed they were going to have llamafile bundled and went looking through to see what model they were using by default.

100%, trying a bit of an experiment like this(similar in that I mostly just care about playing around with different agents, techniques etc.) it has built out literally hundreds of tests. Dozens of which were almost pointless as it decided to mock apis. When the number of failed tests exceeded 40 it just started disabling tests.

And most of the big players now have some kind of browser or bowser agent that they could just leverage to gather training data from locked down sources.

^This 100%. Junior SWE here. Agentic coding has kinda felt like a promotion for me. I code less by hand and spend more time on the actual engineering side of things. There’s hype in both directions though. I don’t AI is replacing me anytime soon(fingers crossed), but it’s already way more useful than the skeptics give it credit for. Like most things the truth’s somewhere in the middle.

That’s why I like how OC handles sharing sessions https://opencode.ai/docs/share/

Wish other tools would copy this functionality(and maybe expand it so colleagues can pick up on sessions I share)

In terms of training your own models, is there enough COBOL available for training or are you going to have to convince your customers to let you train on their data (do you think banks would push back against that?)

Yea, that’s something that stuck out to me as well. Seems like auto save and version control could be done in a minimalist manner whist improving QOL

Actually seems like a really cool product, not sure if it would be that useful for any serious reading due to the small size but probably useful for displaying a calendar and other widgets.

Out of interest, what reader is this? Sounds interesting

“ These endpoints are focused on higher tool-calling accuracy by routing requests to providers that demonstrate measurably better tool calling performance.

Exacto endpoints are available for: - Kimi K2 - DeepSeek v3.1 Terminus - GLM 4.6 - GPT‑OSS 120B - Qwen3 Coder ”

Agreed, “ Popularity as a better indicator”. Hypothetically you could look at popularity over time to filter out viral rot content and work out if people feel the content is useful.

Ryanair is probably a better comparison.

We need a benchmark to test this, give an agent access to a browser and card details and set it free.

That animated ascii art on your site is really cool, something I will definitely have to keep in mind for documenting some of my projects in future.

Haven’t looked into them much but I thought the Chinese labs had released some for this kind of thing

Thought the same but from what I can see those are somewhat uncommon, presumably you are unable to affix them to the board as securely as the typical connectors

Very cool product, just a heads up the footer links don’t seem to actually work on that link(I am using safari mobile)

I think it’s the logical move to ensure Ollama can continue to fund development. I think you will probably end up having to add more tiers or some way for users to buy more credits/gpu time. See anthropic’s recent move with Claude code due to the usage of a number of 24/7 users.

Is this the stealth models horizon alpha and beta? I was generally impressed with them(although I really only used it in chats rather than any code tasks). In terms of chat I increasingly see very little difference between the current SOTA closed models and their open weight counterparts.

Was it not curl https://arstechnica.com/gadgets/2025/05/open-source-project-...

I have used https://www.photoprism.app/ and have found the face recognition to work quite well.

Just watched the demo video and thought it is a very interesting approach to development, I will definitely be following this project. Good Luck.

Nice project! Looks good and seems like something I’d genuinely want to use day to day.