Hey everyone—I just grabbed the new DEVONthink 4 beta, which launched this year. Surprisingly, I haven’t seen much buzz online yet, which feels like a missed opportunity considering how powerful and useful this tool is.
I jumped in to see their integration with LLM models, and they definitely deliver on that front. The enhanced OCR features also look promising—I’m particularly excited to automate OCR workflows for my handwritten notes.
I’ll report back in a few weeks after putting it through its paces. If you’re also exploring the beta, please share your experiences!
I’ve been using it since the first day, and so far haven’t had any major problems at all, and only a couple of minor bugs, one of which has already been fixed (we’re on Beta 3 now). The DEVONthink team, of course, say that it’s a beta and shouldn’t be used for production work, but I’ve been using it full-time and am very happy with it.
I don’t use – or need, or want – any of the AI features, but those who do seem to think highly of the implementation, and even without AI, V4 has many solid improvements worth upgrading for. I’d definitely recommend trying it out.
The payment model changes with this version, to one with which users of Tinderbox will be familiar – you pay for a year’s upgrades but keep all the previous features even if you don’t upgrade. As I’d used the two previous versions for 14 years and only had to pay twice, this seems fair enough to me.
In truth, I’m not a heavy DEVONthink user. Hopefully someone using DEVONthink a bit more can chime in. v4 was in public beta for c.3 months so I’d hope most of the easily found issues were found/addressed. If nothing else, DV4 is certainly stable.
Thank you! I am still on DTP3, and have been for years. Been using DT for approximately 20 years, so lots of stuff in it that I rely on on a daily basis.
Thus your note was very encouraging!
