sciextractai

sciextractai – a lightweight Python package that extracts and structures key scientific insights from complex texts.
It is especially useful for fast‑moving fields like attoscience where researchers need concise, structured summaries of breakthroughs, methods, and theories.

---

📦 Installation

pip install sciextractai

---

🚀 Quick Start

from sciextractai import sciextractai

# Example scientific text (e.g., a recent attoscience paper)
user_input = """
Scientists have generated the shortest light pulse ever recorded, lasting only 3 attoseconds.
This breakthrough opens new possibilities for probing electron dynamics in atoms.
"""

# Call the extractor (defaults to ChatLLM7)
extracted_data = sciextractai(user_input)

print(extracted_data)

Output

A list of strings that match the extraction pattern defined in the package, e.g.:

[
    "Shortest light pulse: 3 attoseconds",
    "Implication: Probing electron dynamics in atoms"
]

---

🛠️ Function Signature

def sciextractai(
    user_input: str,
    api_key: Optional[str] = None,
    llm: Optional[BaseChatModel] = None
) -> List[str]:

Parameter	Type	Description
user_input	str	The scientific text you want to process.
llm	Optional[BaseChatModel]	A LangChain LLM instance to use. If omitted, the package creates a default ChatLLM7 instance.
api_key	Optional[str]	API key for ChatLLM7. If omitted, the function looks for the environment variable LLM7_API_KEY.

---

🔧 Default LLM (ChatLLM7)

If you don’t provide an llm argument, sciextractai will instantiate ChatLLM7 from the langchain_llm7 package:

from langchain_llm7 import ChatLLM7

resolved_llm = ChatLLM7(api_key=api_key, base_url="https://api.llm7.io/v1")

• Installation: pip install langchain-llm7
• Documentation: https://pypi.org/project/langchain-llm7/

---

🌟 Using a Custom LLM

You can pass any LangChain‑compatible LLM (e.g., OpenAI, Anthropic, Google Gemini).

OpenAI

from langchain_openai import ChatOpenAI
from sciextractai import sciextractai

llm = ChatOpenAI(model="gpt-4o-mini")
response = sciextractai(user_input, llm=llm)

Anthropic

from langchain_anthropic import ChatAnthropic
from sciextractai import sciextractai

llm = ChatAnthropic(model="claude-3-haiku-20240307")
response = sciextractai(user_input, llm=llm)

Google Gemini

from langchain_google_genai import ChatGoogleGenerativeAI
from sciextractai import sciextractai

llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash")
response = sciextractai(user_input, llm=llm)

---

🔑 API Key & Rate Limits

• ChatLLM7 free tier provides enough quota for most research usage.
• To use a personal key, set the environment variable LLM7_API_KEY or pass it directly:

response = sciextractai(user_input, api_key="your_own_api_key")

• Obtain a free key by registering at: https://token.llm7.io/

---

🐞 Bugs & Feature Requests

Please raise any issues or feature ideas on the GitHub repository:

Issues: https://github.com/chigwell/sciextractai/issues

---

👤 Author

Eugene Evstafev
📧 Email: hi@euegne.plus
🐙 GitHub: chigwell

---

📄 License

This project is licensed under the MIT License.