Hacker News new | past | comments | ask | show | jobs | submit login
Dify, a visual workflow to build/test LLM applications (github.com/langgenius)
184 points by mountainview 11 days ago | hide | past | favorite | 38 comments





> Dify is licensed under the Apache License 2.0, with the following additional conditions ...

I am totally fine with closed-source/commercial licenses, but please don't do a "Like Apache 2.0 but not really" type of license. It just confuses everyone.

You can pick from SSPL, BSL, Elastic license among others if you don't want to roll out your own.

> 2. As a contributor, you should agree that: a. The producer can adjust the open-source agreement to be more strict or relaxed as deemed necessary. b. Your contributed code may be used for commercial purposes, including but not limited to its cloud business operations.

This is not very contributor-friendly.

You could consider keeping an open-source core, and extensions for paid features.


> This is not very contributor-friendly.

You've put it way too politely IMHO. It's a license designed to attract contributors to help build their product for free, but leaves space for them to just re-license the product and sell it when the opportunity arises. It's not like we haven't seen this play already, and it hurts both contributors and users...


genuine question - out of the closed-source/commercial licenses, which ones are the most developer friendly ?

By commercial licenses, I meant those written by lawyers specifically for the product.

But usually you could just go with something like SSPL - which has sufficient protections for the developer (but not the user).


Wow I've never seen so many fake accounts on a HN post before. So then is it fair to say the Github stars for this project could also perhaps be artificially inflated? This month they started to go exponential: https://github.com/langgenius/dify?tab=readme-ov-file#star-h...

Well, I think a lot of the uptick happened last week because that's when it was published in the Toughtworks Tech Radar for this quarter. The audience is large, presumably larger than HN, and that's how I found out about it and have been toying with it since then. I have no idea what I'm doing, but as far as I can tell, this seems like a legitimate project.

This aligns with what you said:

https://devboard.gitsense.com/langgenius?id=4d1dec9067&r=lan...

I created insights for the last 4 weeks and number of new contributors and stars peaked last week. This project has all the signs of a successfully organically grown project.


never heard of tech radar, interesting thanks!

The AI Girlfriend posts are on a lot of threads today, not just this one. But you never know.

Yup, this reeks of sock puppets

Had someone mention it to me yesterday in an organic convo so there are definitely people out there using it (seemingly happily).

Haven’t tried it yet - still evaluating autogen.


> Github stars for this project could also perhaps be artificially inflated?

Maybe, but I don't think so. The number of people engaging with the project is what I would expect from the number of stars received.

https://devboard.gitsense.com/langgenius?r=langgenius%2Fdify...

I've seen projects with 5,000+ stars in a month and only have 20-30 people interacting with the project.

Full disclosure: This is my tool


"5.5k stargazzers"

fyi typo


How does it compare with MagickML?

https://github.com/Oneirocom/Magick


Very slick and potentially very powerful. After a few minutes playing with it, I have a few recommendations:

- Variables should have more types, like an array of objects

- Prompting should incorporate Jinja2/Nunjucks

- For every prompt, I should be able to create many different test examples, along with an answer key, and measure how well it does across many tests

- It should auto-save. I did a lot of prompting work and then clicked another icon. When I came back, all my work was gone. (In fact, I don't see where to save at all! Maybe I'm just missing it.)


How does this compare to n8n? https://github.com/n8n-io/n8n

I don't think they have the same function

What kind of people are using this AI dev platforms?

When do they become better then just rolling your own custom code?


I work on Dify.

what we're trying do with Dify currently is to let people put together prototypes quicker and either get to production or fail at a faster rate.

we've seen it being helpful for non-technical folks to collaborate on a project well (e.g. importing documents for knowledge base, creating no-code workflow apps, etc)


Where are all these workflow apps getting their ui from? Is there some JavaScript library for boxes connected with lines?

Yes! https://reactflow.dev/ is what I’ve been using, has been great.

Cool! Thanks for sharing. It explains why this type of ui is being used so often. Building this from scratch isn’t the hardest thing in the world, but I can imagine it is somewhat of a challenge to get just right

Dify looks super powerful! Always nice to see a React Flow app in the wild :)

All these spam comments have pushed this to the top feed

Sorry about all these spam comments lol

I flagged a few, then realized the problem is way bigger than me

> https://github.com/langgenius/dify/blob/main/LICENSE

everyone is apparently a license pioneer


This is exactly the type of license Redis tried to move to when people got upset and forked it twice.

"Essentially Apache with restrictions"


Yeah, that's kind of weird - Apache, except totally not Apache. They probably meant to do a source-available license with free non-commercial use allowing inbound contributions - makes sense for a startup, but... please get a lawyer and clean it up :).

team member here - totally see where you're coming from, we'll be relaxing this soon.

"AI means the end of coding" didn't age well.

It turns out to get the most out of LLMs you need to program them.


sql used to be the end of coding to retrieve data as well. the circle continues

LLMs mean now you can write simple imperative code to perform NLP tasks.

wtf is up with all these bots ?

it's ironic they decided to do this on a post about LLMs. Are they feeling threatened that LLMs are taking their jobs ?


My theory is, there is an AI frameworks turf war going on and the bots are to ensure discussion is impossible or extremely difficult.

I’m seeing them in a bunch of other threads. This is the first time I have seen HN being spammed like this.

Yeah super weird.

[flagged]


@dang (I hope it's okay to just that you, my nh reader doesn't seem to have a report button)



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: