This directory contains example scripts demonstrating how to use deepxiv-sdk.

• quickstart.py: The fastest way to get started - minimal example to get up and running

• example_reader.py: Basic usage of the Reader class for accessing arXiv papers

  • Searching for papers
  • Getting paper metadata
  • Reading sections
  • Getting previews and full content

• example_agent.py: Basic usage of the Agent class for intelligent paper analysis

  • Simple queries
  • Follow-up questions with context
  • Detailed paper analysis
  • Using different LLM providers (OpenAI, DeepSeek, OpenRouter)

• example_advanced.py: Advanced Agent usage patterns
  • Literature reviews
  • Methodology comparisons
  • Concept explanations
  • Research trend analysis
  • Finding related work
  • Multi-query interactive sessions

1. Set up your API tokens as environment variables:

# deepxiv CLI auto-registers DEEPXIV_TOKEN on first use and saves it to ~/.env
# You can also set it manually if needed:
export DEEPXIV_TOKEN="your_deepxiv_token"
export OPENAI_API_KEY="your_openai_key"
# Or for DeepSeek:
export DEEPSEEK_API_KEY="your_deepseek_key"

2. Install the package:

pip install deepxiv-sdk[all]

3. Run an example:

python examples/quickstart.py
python examples/example_reader.py
python examples/example_agent.py
python examples/example_advanced.py

All examples can be customized by:

• Changing the model (e.g., "gpt-4", "gpt-3.5-turbo", "deepseek-chat")
• Adjusting parameters (temperature, max_tokens, max_llm_calls)
• Using different LLM providers (OpenAI, DeepSeek, OpenRouter, etc.)
• Modifying the queries to suit your research needs

• Use print_process=True to see the agent's reasoning steps
• Use stream=True for real-time response streaming
• Start with simpler queries and build up to more complex ones
• The agent maintains context across queries, so you can ask follow-up questions
• Use agent.reset_papers() to start fresh on a new topic