I Built The Same App with Every LLM

Click-bait thumbnail fulled by AI's dying hype

Brother, you are the last person I thought would hop on this trend

I think Kimi is ranked too high for an unplayable game, compared to Gemini, which “worked” the first time and is somewhat playable. Include GLM 5.2 in your next test, it would be interesting to see

Can you give an idea of cost? If one model charged twice and was slightly better...

can we get all the prompts and assets you used for mario kart?

I have so many nostalgic memories of watching your pygame coding videos when I was a freshman in high school and now I am interning as a software engineer at a tech company. I feel like i have come full circle thank you for the content!

I like these head to head comparisons, thanks Tim! Good video

Start reviewing your code with CodeRabbit: https://coderabbit.link/techwithtim

Which framework was used to create the interactive initial problem definition and build comparison visualization? Is the code available? I would like to use it for my presentations

Love this. Nice to see someone being a bit more scientific! Best at certain tasks like planning, following a plan, reviewing and polishing ui. Best at specific things like good abstraction, clean code, refactoring, following a codebases style. Best at using subagents, bonus: make your own sub agents with pi based on all the above findings