Dify API

make.fan logomark

Dify API International Version

Dify provides a "Backend-as-a-Service" API, which brings many benefits to AI application developers. This approach enables developers to use the power of large language models (LLMs) directly in front-end applications without facing the complexity of back-end architecture and deployment processes.

Go to install

Please note: If your host is us2 or eu1, be sure to replace the prefix part in the URL with the corresponding region to ensure it is consistent with your host so that it can be installed correctly. It also supports using the Dify open source code downloaded from Github and running on your own server.

Dify is an open source large language model (LLM) application development platform. It combines the concepts of Backend as Service (Backend as Service) and LLMOps, allowing developers to quickly build production-level generative AI applications. Even if you are a non-technical person, you can participate in the definition and data operation process of AI applications.

Because Dify has built-in key technology stacks required to build LLM applications, including support for hundreds of models, intuitive Prompt orchestration interface, high-quality RAG engine, robust Agent framework, flexible process orchestration, and also provides A set of easy-to-use interfaces and APIs. This saves developers a lot of time reinventing the wheel, allowing them to focus on innovation and business needs.

Why use Dify?

You can probably think of a development library (Library) like LangChain as a toolbox with a hammer and nails. In contrast, Dify provides a complete solution that is closer to production needs. Dify is like a set of scaffolding and has undergone sophisticated engineering design and software testing.

Importantly, Dify is open source and is built by a dedicated full-time team and the community. You can self-deploy capabilities like Assistants APIs and GPTs based on any model, with flexibility and security while maintaining full control of your data.

What can Dify do?

Startup Accelerator

Quickly turn your AI application ideas into reality, regardless of success or failure, you need to accelerate. In the real world, dozens of teams have obtained investment through Dify to build MVP (minimum viable product) or won customer orders through POC (proof of concept).

LLM Integration

Integrate LLM into existing business, enhance the capabilities of existing applications by introducing LLM, access Dify's RESTful API to decouple prompts from business code, and track data, costs and usage in Dify's management interface, continuously Improve application performance.

Enterprise-grade infrastructure

As an enterprise-level LLM infrastructure, some banks and large Internet companies are deploying Dify as an LLM gateway within the enterprise to accelerate the promotion of GenAI technology within the enterprise and achieve centralized supervision.

Explore the potential of LLM

Explore the capabilities of LLM. Even if you are a technology enthusiast, you can easily practice prompt engineering and agent technology through Dify. Before the launch of GPTs, more than 60,000 developers had created their first applications on Dify.

Introduction to operating functions

(I.e.

Agent mode selection

1
Chat Messages Agent Model Action

Supports Chat Messages Agent mode in Smart4A AI Agent Server.
Support file import.

2
Chat Messages Assistant Model Action

Supports Chat Messages Assistant mode in Smart4A AI Agent Server.
Support parameter and file input.

3
Workflow Model Action

Supports the Workflow Model mode in Smart4A AI Agent Server.
Support parameter and file input.

(I.e.

Matching tools

1
File upload

Upload files (currently only image formats are supported) for use when sending messages to achieve multi-modal understanding of images and text. Supported formats include png, jpg, jpeg, webp, and gif. Uploaded files are only available to the current user.

2
Next Suggested Questions

To get suggestions for next questions for the current message, this function module will provide three suggestion options.

3
Get Conversation History Messages

Return historical chat records in the form of column loading, and return the latest 20 messages, that is, sorted in reverse order.

4
Delete Conversation

After deleting the conversation record, the API will not be able to retrieve the conversation data because the ID no longer exists. However, in the [Log and Annotation] interface on the host side, the conversation records that have occurred will still be retained.

(I.e.

Voice conversion

1Text to Audio

Text to speech. This is a technology that converts written text into speech output. It is usually used in voice assistants, accessibility technology, voice navigation and other scenarios, allowing the device to read text content in the form of human voice.

2Speech to Text

"Speech-to-text" is a technology that converts voice signals into corresponding text content. This technology is usually used in speech recognition systems, such as voice assistants, subtitle generation, automatic recording, etc., to help users convert speech content into readable text.

3Speech to Speech

STS (audio to audio) technology allows the system to understand the voice and respond directly to the voice, creating a complete two-way voice interaction experience. It enables people to understand speech, speak and respond, and upgrades from one-way voice conversion to two-way intelligent voice interaction, making communication more intuitive and smooth.

4ClearJson

To prevent Dify from including unnecessary text in its output, perform json data cleaning operations.

Official API authorization: Paid application required
Kit installation serial number: Need to pay to purchase

Recommended offers

annual payment plan

Pay in USD Here

⚠️ US dollar payment does not support Taiwanese credit cards

monthly payment plan

Pay in USD Here

⚠️ US dollar payment does not support Taiwanese credit cards

To use this application module, you do not need to apply for an API Token or perform OAuth 2 authentication. You can obtain the module's authentication serial number by selecting an applicable fee plan and completing the payment.

Points to note

This kit is authorized by serial number. Once the serial number is issued, it cannot be recovered, so no refund service can be provided.
Please be sure to confirm the required authorization period before purchasing to ensure your rights.

tips

Official serial number

Most APIs provide their own official authorization API Token or perform OAuth 2 authentication, and the official will set the service price according to the usage. Therefore, you need to apply to the official before it can be issued.

API for Make GUI installation serial number

The serial number issued by make.fan, this API is mainly used for general program applications, and cannot be used directly through the GUI on the make platform. Smart4A Team puts this service on the make platform in the form of GUI, which is convenient for users to use directly. To use this module, you need to register with Smart4A. For more details, please refer to [Pricing].

Graphical User Interface (GUI)

An application is a software interface displayed on the screen with which users can interact.


The price here is for the GUI version, not the official price of the API (see "How to Buy" for instructions).
This application uses rental system, divided into two payment methods: monthly payment and annual payment. The charging price of this application module is as follows:

rental system

Highly recommended

annual payment plan

NT $ 2,500 /year
89.90 USD/Yearly

monthly payment plan

NT $ 250 /month
8.99 USD/Monthly

Notes

The billing method of this mall is rental system And counting system Two modes.

1
rental system

This is the monthly and annual payment mechanism that uses the SaaS model. Users can enjoy the features of most SaaS models without points or times restrictions through regular monthly payments (some services may have upper limits or be differentiated by versions). Annual payment plans are usually priced based on a ten-month total, which means two extra months of usage, but not all plans are like this.

2
counting system

It is a point deduction mode, where users need to purchase a certain amount of points, and the points will be gradually deducted based on actual usage.