OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

AppBuilder-SDK

AppBuilder-SDK is a Python client SDK for Baidu Qianfan AppBuilder that lets developers call models and components, manage knowledge bases, compose AI workflows and build RAG applications.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

AppBuilder-SDK — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when developing a Python application that needs Qianfan model access or when composing AppBuilder components and workflow stages into an AI or RAG application.

Who it is for

  • Python developers building applications with models available through Baidu Qianfan.
  • Teams assembling RAG applications from document-processing, embedding, retrieval and answer-generation components.

Documented uses

  • Call a Qianfan model with a prompt template and chosen parameters.
  • Build a RAG pipeline that parses and chunks documents, creates embeddings and retrieves relevant material for answer generation.
  • Manage knowledge-base documents and chunks, or compose workflows using Message, Component and AgentRuntime abstractions.
  • Connect an application to published AppBuilder applications and Baidu ecosystem components.
Before you choose

Requirements & limits

The Python installation instructions require Python 3.9 or later. The README quick start uses a personal token and a model the developer has permission to access through Qianfan. Certain capability-component calls require claiming a free trial quota. The repository is public under Apache-2.0.

  • The SDK is a client for the Baidu Qianfan AppBuilder platform, not a standalone model host.
  • The documented Python operating systems are Linux, Windows and macOS.
  • The README describes more than 40 Baidu-ecosystem components, with a free-trial-quota prerequisite for component calls.
  • AgentRuntime deployment options described in the repository include Flask with gunicorn and a Chainlit conversation interface.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the AppBuilder SDK and invoke a Qianfan model

Install the Python SDK and use its Playground component to request a model response with a prompt template.

Before you start

  • Python 3.9 or later.
  • An AppBuilder personal token, which the example places in the APPBUILDER_TOKEN environment variable.
  • Access to a model on the Baidu Qianfan large-model platform.

Steps

  1. Install or upgrade the SDK with the command below.

    Documented command
    python3 -m pip install --upgrade appbuilder-sdk

  2. Set APPBUILDER_TOKEN to your personal token in the Python environment.

  3. Create an appbuilder.Message with values for the prompt template, then initialize appbuilder.Playground with the template and an available model.

  4. Call the Playground with the message and stream=True; iterate over output.content to display streamed response content. The example also prints output.model_dump_json(indent=4) after streaming.

Expected outcome

The example displays streamed model response text and then prints the complete response object, including token-usage information.

Guide limitations

  • The SDK requires Python 3.9 or later.
  • The example requires a personal token and a model the user is permitted to call.
  • The README's displayed model name is an example; model availability and permissions are not established for every account.
Documentation 1 ↗
Source-based practical guide

Call the Baidu Search-enhanced RAG component

Use the SDK's RAG with Baidu Search Pro component to submit a message with an instruction and print its result.

Before you start

  • Python 3.9 or later with the AppBuilder SDK installed.
  • An AppBuilder personal token set in APPBUILDER_TOKEN.
  • The README says to obtain free trial quota before calling capability components.

Steps

  1. Set APPBUILDER_TOKEN to your personal token in the Python environment.

  2. Initialize appbuilder.RagWithBaiduSearchPro with a model name.

  3. Wrap the query in appbuilder.Message and call the component's run method with that message and an instruction wrapped in appbuilder.Message.

  4. Print the returned result using result.model_dump_json(indent=4).

Expected outcome

The example prints a structured result containing an answer to the submitted query.

Guide limitations

  • The README states that capability components require applying for free trial quota before use.
  • The model shown is an example; the source does not establish that it is available to every user.
  • The example describes the component as combining Baidu Search technology with ERNIE semantic understanding; results depend on the service and model.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
9 October 2026
Original website title
baidubce/app-builder · Official homepage title · retained in its source language
How we review sourcesSuggest a correction