- Very slow to start up and TTFT is terrible. lmpify hits within 100ms and doesn't need to wait for pageload to start seeing result, result is already starting in DO as soon as you submit.
- Claude doesn't support URLs out of the box. When i create a new app/worker, i usually add URLs as source of truth for context . with claude it was still a hassle to copy/paste afterwards. now, this is seamless and fast.
- Claude can render HTML and React. I don't use React so I don't care. I use HTML. Claude HTML renders are super limited, as they can't run scripts. LMPIFY HTML renders render scripts and anything else, and can be easily opened in fullscreen.
- Claude is generally slow, buggy, and unreliable on my machine. LMPIFY is snappy / fast
- It's hard to share something with someone else in claude, requires several clicks. LMPIFY is optimised for sharing
- LMPIFY incentivizes people to edit the prompt rather than reply, which usually gives better and less lengthy results as token windows become shorter from it. It also incentivizes people to reduce tokens. Claude does NOT do this
I think this is my main list of why I built it. I don't think it'd replace full LLM use for most, but maybe for sharing and when the edit-flow preferred. Maybe I'm the only one that needs it. IDK. Please give it a try (got some free tokens for everyone) and let me know if you find it useful for casual use!