Skip to content

Chat with Your Data

Lumen is an open-source AI agent framework for interactive data analysis. Ask questions about your data—Lumen generates SQL, runs it, and shows you charts. No coding needed.

What Makes Lumen Different

Feature Lumen Other Tools
Your LLM Use any LLM (yours or cloud) Locked to one provider
Your data Any source: local, DB, API, lake Limited connectors
Extend it Build agents, tools, visualizations Limited customization
Open source 100% open, no telemetry Often proprietary
Declarative YAML + Python, easily reproducible Often code-only

Get Started in 30 Seconds

Using one of the methods below, start Lumen with a sample dataset, open https://localhost:5006, and start asking questions!

Command line:

pip install 'lumen[ai-openai]'  # or see Installation guide for other LLMs
lumen-ai serve https://datasets.holoviz.org/penguins/v1/penguins.csv

Python:

Create app.py:

from lumen.ai import ExplorerUI
ui = ExplorerUI(data='https://datasets.holoviz.org/penguins/v1/penguins.csv')
ui.servable()

Then run:

panel serve app.py

Tip

You may need to specify an API key for your chosen LLM provider, e.g. export OPENAI_API_KEY=your_key_here for OpenAI. See LLM Providers for details.

How It Works

You don't need to write SQL or Python; just ask a question, e.g.

→ Which islands have the most penguins? Plot as a horizontal bar chart.

Lumen creates a plan:

- [ ] Query the penguin dataset to aggregate the total number of penguins by island. Provide the pipeline and with the aggregated data.
- [ ] Using the aggregated penguin count by island, create a horizontal bar chart showing islands on the y-axis and penguin counts on the x-axis.

From the plan, Lumen generates SQL to query the data:

SELECT "island", COUNT(*) AS "penguin_count" FROM penguins GROUP BY "island" ORDER BY "penguin_count" DESC
island penguin_count
Biscoe 168
Dream 124
Torgersen 52

Then Lumen creates a Vega-Lite spec to visualize the results:

$schema: https://vega.github.io/schema/vega-lite/v5.json
  data:
    name: penguin_count_by_island
  height: container
  layer:
  - encoding:
      color:
        value: '#4682b4'
      x:
        axis:
          title: Penguin Count
        field: penguin_count
        type: quantitative
      y:
        axis:
          title: Island
        field: island
        sort: -x
        type: nominal
    mark: bar
  title:
    anchor: start
    fontSize: 20
    subtitle: Biscoe island has the highest penguin count, followed by Dream and Torgersen
    subtitleColor: '#666666'
    subtitleFontSize: 16
    text: Penguin Counts by Island
  width: container

Community & Support