Structured Data Extracted Using AI
Tips for Better Results
- Use specific field descriptions to guide the AI accurately
- Start with more fields and remove unnecessary ones later
- If results are inaccurate, try rephrasing your field descriptions
- For multiple text blocks , separate each with a double line break
- Each text block should contain complete information for all fields you want to extract
Technical Note
To ensure timely results (since this is hosted on a CPU), we utilize
gpt-5-mini-2025-08-07
for this demo.
For
complete privacy
, consider using local open-weight models via
chat_ollama()
from the
ellmer
library, which connects to
Ollama
for running models locally on your machine.
Alternatively, for higher accuracy requirements, users can leverage more performant closed models (e.g.,
gpt-5.2-2025-12-11
,
claude-opus-4-5-20251101
, or
gemini-3-pro-preview
) depending on the application needs.
Note that Claude models require
chat_anthropic()
and Gemini models require
chat_google_gemini()
, each with their respective API keys configured in the R environment.
This demo uses
chat_openai()
only and does not provide an option to change the gpt-5-mini-2025-08-07 model.