• 1 Post
  • 10 Comments
Joined 2 years ago
cake
Cake day: February 6th, 2025

help-circle
  • OpenRouter let’s you buy ‘ai access’ - different models (agents like Claude or GPT is specific forms like Fable or Sol) from different routers (providers aka data centers).

    So you get an API key and save it in your setup (IDE, coding app, CLI env whatever) and then you have a library of all the models available pay as you go. Prices per model are competitive if they’re available from multiple providers or at the market rate from folks like Anthropic and OpenAI.

    For AI service at home I switched from Copilot, a MS subscription with opaque usage pricing to OpenRouter so I can pick / route work to cheap or free models available on OR while still using my preferred models and having access to everything under the sun including the latest versions. It’s pretty great honestly.

    OR is also a good place to look if you just want some info about AI models and providers. Their benchmark, popularity and domain information makes the models make sense and highlights real world usage patterns, i.e. DeepSeek v4 and GLM 5 are serious competition to the proprietary models like Claude with pricing 10-100x less. Presumably OR is just taking a cut as the middleman to pay for the service.

    And in case you read this far and you’re wondering, I’m using it with Zed and working on a custom fork to improve the agent handling and routing to optimize use of such services. Enjoy the AI, spare me any generic ai outrage.


  • Here here. This place doesn’t have much rational thoughts on AI.

    I’m (a little) surprised how group think this place is on the topic - y’all are a serious echo chamber. This place is like rainbow flags and kill the bots central. Any pro AI counterpoint mentions bring pitch forks and hate. That’s not a good look.

    Maybe I’m an AI bro myself, but I’m also an experienced professional using it to do real work (and no, not taking any jobs, building new tools for a team that needs them). Linus is a real expert. But why listen to experience and expertise when you have an opinion and feel justified. It’s ok to be afraid, but when scary change comes wise folks listen and learn.

    Here we don’t worry about facts and reality, just get your outrage revved up. Social media is social media - and that’s scarier than AI, because it uses people to make the garbage.



  • So you teach something different from the mainstream - cool. But don’t be surprised when students or AI cite others or take an alternative position.

    I’m no economist but I had plenty of professors who wanted their particular theories reflected back at them.

    While students should do their own reading and writing, using ‘they write mainstream ideas’ as a litmus test for writing and AI seems parochial. And I’d go so far as to say AI is becoming a standard research and writing aid like the web word processors, and you sound like teachers who ranted against those too.

    Have you engaged either group in open discussion about comparing the perspectives you’re challenging? I bet several modern models could debate your philosophy on merit, I wonder if you ever gave students the chance.






  • Thanks for sharing, interesting read and questions. Surely you’ll be down voted here for anything with AI… But c’est la vie.

    Ive been doing coding projects in VS code which uses GPT, Claude and Gemini. Woe are the days when my credits are used and only GPT 4.1 is available. Claudes ability to research and architect multi step software solutions is very, very good and it rarely makes messes or spins tires compared to older models from just a few months ago. This is precisely what converted me to ‘whoa - ai’ which is adjacent to ‘pro ai’.

    Lately I’ve been experimenting with customizing Gemini via instructions which include a link to a drive folder of md files with specific instructions for different agent tasks, such as performing specific market analysis, doing a news roundup with a specific list of topics and omitting prior reviewed items, etc. The files allow for both complex instructions or lists, as well as some chance to construct memory via logging. Results are a mixed bag, lots of additional function created, lots of mixed results.

    Have you considered any tests of more complexity? Something like ‘write a program that…’ I think what will differentiate these models going forward is some have architect capabilities, strategy, insight, decision making, where others are agents - they do specific tasks well but have limits. With that model, the ai architect and it’s ai agents need to work as a team to complete a multi step task.


  • T3CHT @sh.itjust.workstoMemes@lemmy.mlEspecially with their low pay
    link
    fedilink
    English
    arrow-up
    6
    arrow-down
    1
    ·
    10 months ago

    The underfunded school, or its expensive insurance will pay, raising rates and eroding the service of public education further.

    I found out here in CA public schools are paying several percent of total budget to cover liability from the past where kids were harmed.

    These are not for profit, these are not businesses. When they pay, the taxpayers pay, and the kids lose service.

    But whatever, she got hers. Big smile.