Claude Haiku 4.5 results with 60% fewer tokens. Sounds good, but they don't list token costs.
MAI-Code-1-Flash
141–150 of 297 posts
Re: MAI-Code-1-Flash
#142Re: MAI-Code-1-Flash
#143What's with the lack of Microsoft design language on the website? It's painfully obvious they're trying to emulate Anthropic's style here and it looks tacky.
This is needlessly embarrassing, seems like a small thing, but it makes them look... desperate?
Re: MAI-Code-1-Flash
#144wtf are they doing to the scroll on that page
Re: MAI-Code-1-Flash
#145“ Build for developers, not benchmarks” is the worst marketing shot I ever heard
It claims that, then promptly proceeds to showcase a bunch of benchmarks.
Re: MAI-Code-1-Flash
#146Earlier quoted context omitted.
What’s your prompt for this, the way you described it made it seem like there’s a generalizable way I can go about this. I just rely on a testing pipeline instead so can’t think of why I would need to proactively find holes where tests haven’t already done that for me.
I use similar workflow. Here is my refactoring and code quality prompt that I regularly run: Perform a thorough analysis of the project (the code and the documentation). - Explore the project, go over all important files one by one and look for any mistakes or possible bugs. - Look for refactoring opportunities and ways to improve code quality and organization. - Identify any potential cruft/bloat, to ensure our code…
Thank you!
Re: MAI-Code-1-Flash
#147Earlier quoted context omitted.
What’s your prompt for this, the way you described it made it seem like there’s a generalizable way I can go about this. I just rely on a testing pipeline instead so can’t think of why I would need to proactively find holes where tests haven’t already done that for me.
tests will not find inconsistent naming, duplicate functions, scenarios you have not thought about testing I use quite plain prompts, nothing fancy: > go over the tests and do a code review, focusing on how well they test inventory management, planner and controller. maybe some tests need to be deleted, maybe other tests need to be added. the end goal should be good coverage of the core features. > do a code review,…
Noted, thank you! Appreciate it
Re: MAI-Code-1-Flash
#148What's with the lack of Microsoft design language on the website? It's painfully obvious they're trying to emulate Anthropic's style here and it looks tacky.
Brand guidelines and web design pretty much don't exist any more as far as I can tell. Gotta get it out yesteday, and the only way to do that is vibe coding, styling be damned.
Re: MAI-Code-1-Flash
#149Related ongoing thread:
MAI-Thinking-1 - https://news.ycombinator.com/item?id=48374362 - June 2026 (64 comments)
Re: MAI-Code-1-Flash
#150It's a start and I welcome competition but I don't think I ever used small cloud models like Haiku 4.5. They are cute but for serious coding they tend to waste your expensive time. And this certainly wont bring me back to GitHub Copilot which I cancelled yesterday. GitHub Copilot had competitive pricing until yesterday when they changed from per-request to one of the most expensive per-token quotas. Seriously, take a…
Haiku does quite well if given a detailed plan. That means much more detail than you otherwise would, but you can still save over e.g. having Opus or Sonnet do everything by having them expand their initial plans into more specific levels of detail and feed it to Haiku (or similar level models).
I personally wouldn't use models that class directly, though - I'd use them in a harness as a "backend" for more capable models. And Haiku itself, as opposed to other smaller models, is still expensive.