Using with MindSearch - Starrise AI

1. Get Your API Key

Open the Starrise AI Console and register an account (or log in if you already have one). After registering, go to API Keys, create a new API key, click to copy it, and save it for later use.

2. Deploy MindSearch

After cloning MindSearch locally and installing the required dependencies (see the MindSearch README), edit /path/to/MindSearch/mindsearch/models.py and add the configuration for calling the Starrise AI API:

starrise_config = dict(
    type=GPTAPI,
    model_type='gpt-4o',
    key=os.environ.get('ANYFAST_API_KEY', 'YOUR ANYFAST API KEY'),
    openai_api_base='https://ai.alad.com/v1/chat/completions',
    meta_template=[
        dict(role='system', api_role='system'),
        dict(role='user', api_role='user'),
        dict(role='assistant', api_role='assistant'),
        dict(role='environment', api_role='system')
    ],
    top_p=0.8,
    top_k=1,
    temperature=0,
    max_new_tokens=8192,
    repetition_penalty=1.02,
    stop_words=['<|im_end|>']
)

After adding this configuration, you can run the relevant commands to start MindSearch. Start the backend:

# Specify the Starrise AI API key
export ANYFAST_API_KEY=<the key you copied above>

# Start
python -m mindsearch.app --lang en --model_format starrise_config --search_engine DuckDuckGoSearch

Start the frontend (using Gradio as an example):

python frontend/mindsearch_gradio.py

3. Deploy to HuggingFace Space

You can also choose to deploy to HuggingFace Space:

Create a new Space at huggingface.co/new-space with the following settings: Gradio Template: Blank, Hardware: CPU basic.
After creation, go to Settings to set the API key.
Upload the MindSearch directory, requirements.txt, and app.py.