You can detect whether Matplotlib is running in interactive mode or whether your environment can show plots, and then either display or save the figure.

if is_interactive():
    plt.show()
else:
    plt.savefig("plot.png")

Example Code

import matplotlib
import matplotlib.pyplot as plt
import sys

# Detect if running interactively
def is_interactive():
    return hasattr(sys, 'ps1') or sys.flags.interactive or matplotlib.is_interactive()

# Example plot
plt.plot([1, 2, 3], [4, 9, 16])
plt.title("Example Plot")

# Show or save depending on mode
if is_interactive():
    plt.show()
else:
    plt.savefig("plot.png")
    print("Plot saved to plot.png")

Explanation

• hasattr(sys, 'ps1') → True in an interactive terminal (e.g. Python REPL or IPython).
• sys.flags.interactive → True if Python was launched with -i.
• matplotlib.is_interactive() → True if matplotlib’s interactive mode is on (plt.ion()).
• plt.show() → Opens a window if possible.
• plt.savefig() → Writes to a file if no GUI is available or script is non-interactive.

Alternative (robust script version)

If you know your code will often run headless (like in servers or CI):

import matplotlib
import sys

if not hasattr(sys, 'ps1'):
    matplotlib.use('Agg')  # Non-interactive backend

import matplotlib.pyplot as plt

plt.plot([1, 2, 3], [4, 9, 16])
plt.title("Server-safe Plot")

if hasattr(sys, 'ps1'):
    plt.show()
else:
    plt.savefig("plot.png")

This explicitly switches Matplotlib to a non-GUI backend (Agg) when not in interactive mode — ensuring it won’t crash on systems without a display.