datialog 1.3.4

Natural AI Data Explorer — analyze datasets in plain language, locally and privately

Datialog — Natural AI Data Explorer

Analyze datasets in plain language using AI. No code required. Your data never leaves your machine.

Installation

pip install datialog

Usage

# Launch (opens browser + system tray icon)
datialog

# Custom port
datialog --port 9090

# Terminal mode (no system tray)
datialog --no-tray

Or from Python:

import datialog
datialog.launch()

Features

• Natural language — ask questions in Spanish or English
• 100% local — data never leaves your machine with Ollama
• Multi-backend — Ollama, Groq, OpenAI, Anthropic, LM Studio
• Multiple formats — CSV, Excel, Stata, SAS, Parquet, JSON
• Auto EDA — exploratory analysis with one click
• Python console — integrated code console
• Checkpoint & Undo — revert transformations
• System tray — background mode with tray icon

Supported formats

Format	Extension
CSV	.csv
Excel	.xlsx, .xls
Stata	.dta
SAS	.sas7bdat
Parquet	.parquet
JSON	.json

Requirements

• Python 3.10+
• For local models: Ollama
• For cloud models: API key from Groq, OpenAI or Anthropic

Author

Ivan Pastor Sanz — datialog.app

Citation

@software{pastor_sanz_2026_datialog,
  author    = {Pastor Sanz, Iván},
  title     = {Datialog — Natural AI Data Explorer},
  year      = {2026},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.20075103},
  url       = {https://doi.org/10.5281/zenodo.20075103}
}

License

CC BY-NC 4.0 — free for personal, academic and research use. Commercial use requires explicit permission from the author.