OpenAI's new $100 tier targets developers hitting Codex limits(thenewstack.io)
14 pointsby Brajeshwar9 hours ago4 comments
  • nialse8 hours ago
    Manufacturing demand. Create the problem. Solve the problem.

    The 2x allowance last month was not of the goodness of their hearts. It’s pumping up the demand AND the revenue simultaneously. The squeeze before an IPO.

    How many of us have looked into moving to the next tier during the last 48 hours? I know I have both professionally and for my private use. Moving 20% of $20 customers to $100 increases the revenue for the group with 1.8x.

    • esperent3 hours ago
      I would agree except that I'm already on the top tier, and there is no next tier. So what I spent yesterday doing when I hit my session limit - within an hour and a half while working on a single feature - was setting up pi.dev and porting over all my skills and hooks.

      I've cancelled my account and when it's finished in a few days I'm done with Claude. Sure, Opus has a slight edge (which might just be familiarity) but it's already clear that pi with Gemini 3.1 pro or codex 5.3 is pretty much close.

    • gruez8 hours ago
      >Manufacturing demand. Create the problem. Solve the problem.

      This only makes sense if openai was cutting the quotas to below the original amount, which so far as I can tell hasn't happened. Otherwise it's just a cynical take where any sort of promotion can be cynically interpreted as "Manufacturing demand. Create the problem. Solve the problem".

    • moralestapia7 hours ago
      Their models also "became" quite chatty in recent weeks.

      Now you spend 2x tokens to achieve the same result because of that.

      Sure, you can tell it to "be succinct" or whatever, but the vast majority of people won't so this definitely increased token consumption overall.

    • nothinkjustai8 hours ago
      I pulled the trigger on a GLM subscription, 5.1 is pretty close to SOTA and as a side benefit you aren’t financially supporting these evil companies.
  • abstractspoon41 minutes ago
    Once enough people have lost the ability to write their own code, they will be fully at the mercy of the price setters. One thing I love about coding as a hobby is that it costs me nothing.
  • denimnerd428 hours ago
    When I sit down and dial in on a serious project at work I easily spend $500-1000+ day of API usage. At home I just hit the limit and give up / cancel my sub.

    Wish I could just go all out one weekend a month. I hardly code outside of work but sometimes there’s a project I have an itch for.

    • chipsrafferty7 hours ago
      You could try writing the code yourself
      • sidrag226 hours ago
        I know some people still underestimate these tools, but this is pretty adjacent to telling someone with a 20mile commute to just walk everyday instead.
        • denimnerd425 hours ago
          I have at least walked 20 miles before in my life. I've never written anything with as much breadth in 20 years of coding until I started using these tools. I also have quite a deep backlog from trying.
    • daymanstep8 hours ago
      Why don't you just buy API tokens instead of using subscription
      • thejazzman5 hours ago
        https://github.com/ryoppippi/ccusage

        I'm at $6.2k over the last 30 days from my $200 codex subscription

        • esperent3 hours ago
          People will keep pointing out that it means the $200 subscription is good value, but to me it just means the Claude API is wildly overpriced.
      • denimnerd428 hours ago
        That’s what Im considering.
    • teaearlgraycold5 hours ago
      What kinds of tasks are you giving to the LLMs? I don’t think I do anything that can rack up costs like that. I can only imagine you’re using lots of instances simultaneously. I’d love to know more and ideally see the deliverables if the code is public, or even just the product.

      I recently talked with a guy that’s pretty smart and is building a good product with a clear market. I understood the idea and encouraged him to ship.

      Then he showed me what he considers his real work, and went off on a madman’s raving presentation of some supposedly hyper scalable revolutionary encrypted block chain dApp agentic operating system. He was building all of this using lots of agents. And I could totally see him spending $500/day on tokens. But I also couldn’t get him to explain the use case. I couldn’t imagine one myself. I’m by default suspicious of large AI bills. People usually only have a short amount of roadmap that’s well thought out and proven. Building faster means you just hit a new bottleneck in product design.

      But I want to learn more and accrue more case studies of AI use in software engineering. Sometimes I hear of some really great software engineering techniques only possible thanks to AI (stuff like running 3 models in parallel optimizing a hot loop, comparing outputs with a rigorous test suite and fuzzing).

      • denimnerd424 hours ago
        We don't spend that much money every day but here's the gist: We have a distributed system that has several components that don't meet the performance requirements of the next uplift we need to do. We need to carefully consider the tradeoffs of things like how to shard a few of the databases, how to rearchitect the ETL flow that comes off the system and is used for analysis. We think of a few approaches and then we get the coding assistants to blast through the end to end development of each approach discovering all the known unknowns and unknown unknowns along the way. Then we can load test each method, profile them, analyze them manually and with the LLM. Then we can pick the solution and take another shot at implementing it with the coding agent, but more carefully and with more oversight with all the things we learned.

        We don't hit those high numbers every day. An average day is $50-100 max.

        As far as home projects. Something like write a GUI desktop or phone application from scratch. The LLM has to reference a lot of code and API docs to figure out what to do and spends a lot of time thinking while debugging. It gets expensive :/