Coze Studio - ByteDance's open-source AI Agent development platform

What is Coze Studio?

Coze Studio is an open-source AI agent development platform from ByteDance, helping developers quickly build, deploy, and manage AI agents. It provides a one-stop development environment, supporting core technologies such as Prompt, RAG, Plugins, and Workflows. Through visual workflow orchestration, developers can create complex AI applications with zero or low code. The platform has a built-in plugin framework that can encapsulate third-party APIs or private capabilities as plugins to extend agent functionality. Coze Studio is developed using Golang and React, adhering to Domain-Driven Design (DDD) principles, making it easy to perform secondary development. It supports local and private deployments, with a low deployment threshold, requiring only a dual-core CPU and 4GB of memory to run, and provides a one-click deployment script.

Main features of Coze Studio

• One-stop development environmentIt provides full-process support from development to deployment, covering core technologies such as Prompt, RAG, Plugin, and Workflow.
• Visual workflow orchestrationIt allows you to quickly build complex workflows by dragging and dropping nodes, supports zero-code or low-code development, and makes it easy for developers to design business logic.
• Intelligent agent construction and managementIt provides robust application templates and orchestration frameworks to support the rapid creation, deployment, and management of AI agents.
• Multi-model supportIt supports the integration of various mainstream large models (such as OpenAI, Volcano Engine, etc.) and provides a unified model interface abstraction.
• Knowledge base managementIt integrates RAG (Retrieval Enhanced Generation) capabilities, supports vector retrieval, allows uploading documents or data, and enables the agent to generate accurate answers based on the knowledge base.
• Plug-in systemIt supports the creation, configuration, and management of plugins, and can encapsulate third-party APIs or private functions to extend the capabilities of intelligent agents.
• Database supportIt integrates database resources and supports data storage and retrieval.
• API and SDK integrationIt provides APIs related to chat and workflow, and supports SDKs such as Python, JavaScript, and Java, making it convenient for secondary development.
• Real-time interactionIt supports real-time chat, speech synthesis, and transcription via WebSocket.

Coze Studio project address

• Github repositoryhttps://github.com/coze-dev/coze-studio

How to use Coze Studio

• Environmental preparation
  • Ensure your computer meets the minimum configuration requirements: dual-core CPU and 4GB of RAM.
  • Install Docker and Docker Compose beforehand, and start the Docker service.
• Get the source codeOpen a terminal or command prompt and run the following command to clone the Coze Studio source code: `git clone https://github.com/coze-dev/coze-studio.git`
• Configuration Model
  • Navigate to the cloned project directory: cd coze-studio
  • Copy the model configuration template file from the template directory: cp backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml backend/conf/model/ark_doubao-seed-1.6.yaml
• Modify configuration file:
  • Enter backend/conf/model Table of contents.
  • Open ark_doubao-seed-1.6.yaml document.
  • Set the following fields:
    • idModel ID, defined by the developer, must be a non-zero integer and globally unique.
    • meta.conn_config.api_keyAPI Key for Model Services.
    • meta.conn_config.model: The model ID of the model service.
• Deploy and start the service
  • Enter docker Copy the environment variable template file from the directory and start the service.
  • The initial deployment may require pulling the image and building a local image, which can be time-consuming. If you see the message "Container coze-server Started", it means the service has started successfully.
• Access PlatformOpen your browser and visit http://localhost:8888/You can then open the front-end page of Coze Studio.
• Registration and LoginOn the Coze Studio login page, register and log in using your email or phone number.
• Creating intelligent agents:After logging in, click the "Create Bot" button in the upper left corner and enter the basic information of the Bot (such as name, avatar, etc.).
• Configure intelligent agents:
  • On the configuration page, you can set the bot's role, skills, and restrictions (hint words).
  • Add official plugins or custom plugins, and configure workflows, knowledge bases, or databases.
  • Interact with the bot on the debug page to test its functionality.
• Use plugins
  • Using plugins in a bot:
    • On the Bot editing page, select the plugin area and click the plus icon to add a plugin.
    • Configure plugin parameters and test plugin functionality in the preview and debug areas.
  • Using plug-in nodes in workflow:
    • Create or select a workflow, and add plugin nodes to the workflow.
    • Configure the input and output parameters and test the plugin functionality.
• Publish applicationAfter completing all configurations, click the "Publish" button in the upper right corner to publish the agent to the Coze app store.

Application scenarios of Coze Studio

• Rapid PrototypingValidation (POC)Developers can quickly build prototypes of AI applications to validate their ideas and concepts.
• AI application implementation in SMEsSmall and medium-sized enterprises can quickly leverage Coze Studio's large model capabilities to solve business problems.
• Private DeploymentFor enterprises with high data security requirements, Coze Studio supports private deployment to ensure that business data does not leak out of the enterprise intranet.
• Internal Q&A RobotUse a knowledge base to build a company SOP intelligent agent, and add logic such as approval, form filling, and email sending through workflow.
• AI Interactive Questionnaire SystemUse workflow to guide users to fill out questionnaires, call plugins to generate data charts, and synchronize the data to the backend BI system.