Deploy JoyAgent‑Genie Locally and Switch to DeepSeek in Minutes
This guide walks you through quickly trying the open‑source JoyAgent‑Genie online, then cloning the repository, running the one‑click start script, and reconfiguring the .env and application.yml files to replace the default GPT‑4.1 model with DeepSeek, all with clear step‑by‑step instructions and screenshots.
Quick Online Experience
JoyAgent‑Genie has been open‑sourced, and you can try it instantly by logging in with a JD.com or JD Cloud account at joyagent-genie.jdcloud.com/.
Local Deployment
For deeper exploration, clone the GitHub repository: git clone https://github.com/jd-opensour… Then start the service with the provided script: sh Genie_start.sh The script launches the application without errors, and opening the displayed URL shows the local Genie interface.
Step 3: Configure the Model
The default model is GPT‑4.1, which is inaccessible from China, so you need to switch to DeepSeek by editing two configuration files.
3.1 Modify the Python .env file
Remove the default OPENAI_API_KEY and OPENAI_BASE_URL, then add your DEEPSEEK_API_KEY, DEFAULT_MODEL, and SERPER_API_KEY as shown in the screenshot.
3.2 Modify the Java configuration
Edit resources/application.yml and update the llm-default section for the DeepSeek model, ensuring max_tokens is set to 8192 to avoid errors.
3.3 One‑click start
Run Genie_start.sh again; the service starts with the DeepSeek model, as demonstrated in the following screenshot.
Summary
We have completed both the quick online trial and the full local deployment of Genie. The results are impressive, producing rich text and charts. Deployment is straightforward, and developers with basic experience can use this open‑source project to build their own Manus services.
JD Tech Talk
Official JD Tech public account delivering best practices and technology innovation.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
