What is our primary use case?
My main use case for Dify Enterprise is building workflows which trigger the LLM calls and perform certain actions.
I can give you a specific example of a workflow I have built using Dify Enterprise: in supply chain management, whenever there is a logistics-related query, an intent agent first identifies whether the query is relevant to logistics or not. After that, there are some calculations according to the context given. In this workflow, we are leveraging the APIs and connectors of Dify Enterprise to accomplish the work via workflows, and I have a sequential workflow for each of the different use cases, which triggers different Dify Enterprise API endpoints.
I have mostly used Dify Enterprise for these sequential workflows only.
How has it helped my organization?
Dify Enterprise has positively impacted my organization as it provides a low-code/no-code platform for building workflows, so there is no need for much experience in building workflows. Roles such as PMO or Project Manager can also tweak it to show around different demos.
Since using Dify Enterprise, I have noticed specific outcomes where it has reduced the time to build workflows and to customize them, as it provides a low-code/no-code platform.
What is most valuable?
The best features Dify Enterprise offers are its very good scaling and security. I appreciate hosting the open-source solution on Dify Enterprise because it is fast and each response does not get bottlenecked.
The security aspect makes a difference in my day-to-day work because Dify Enterprise helps in solving the concurrency problem, which mainly occurs in sequential workflows.
The user experience is good, particularly for people who are used to low-code/no-code platforms, so for them, Dify Enterprise is very intuitive.
Regarding Dify Enterprise's AI capabilities, I think its governance and security are secure, and there are no concerns about governance and security currently affecting me, considering that it is an enterprise level, as all things are managed by the SSO only.
Regarding Dify Enterprise's AI capabilities, I find that for sequential workflows, it is accurate and produces good workflows.
What needs improvement?
I think Dify Enterprise can be improved by adding more connectors and by adding more branching options, such as to trigger a sub-workflow. There should be more options or more conditional options instead of just branching.
I chose 8 out of 10 because I think the connectors can be improved, and after that, the pricing could be on a lower side for the enterprise level.
For how long have I used the solution?
I have been using Dify Enterprise for approximately three months.
What do I think about the stability of the solution?
Dify Enterprise is stable.
What do I think about the scalability of the solution?
Dify Enterprise is scalable because we use the open API in our workflows and it has never failed or had a bottleneck.
How are customer service and support?
I have never contacted customer support as I did not need to contact them or have a use case which needed to be suggested.
Which solution did I use previously and why did I switch?
We did not use a previous solution; Dify Enterprise was our first choice.
What was our ROI?
I have seen a return on investment as employees have more bandwidth available to do different tasks. For different demos, the developers do not need to worry about tweaking the workflows because the PMO themselves or the person who is actually pitching the demo can change the sequential workflows, which reduces some of the time of developers and makes the bandwidth more available.
What's my experience with pricing, setup cost, and licensing?
I am not very sure about the pricing, as it is an enterprise level and the leads, people, or managers handle it. However, from a user's perspective, it is really very helpful, and the setup is easy as well, as it is enterprise.
Which other solutions did I evaluate?
We have not evaluated different options before choosing Dify Enterprise.
What other advice do I have?
My recommendation to others reading reviews on PeerSpot is that Dify Enterprise is a good option if you have a strict problem or strict use case and you know what you are getting into or what you are building using the Dify Enterprise low-code/no-code platform.
Which deployment model are you using for this solution?
Private Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Microsoft Azure