3D Plots

Reviewed & published by Brayan K

Matplotlib is a Python library for creating charts and visualizations — and with its 3D toolkit it can draw surfaces and scatter points across three axes, not just two.

Part of the free Matplotlib course at LearnCodingFast — hands-on lessons with worked examples and the output they print, plus practice exercises and a quick quiz.

In this lesson you'll create a 3D Axes, plot a colorful surface from a coordinate grid, and scatter points in space with x, y, and z labels.

What You'll Learn in This Lesson

• Creating a 3D Axes with projection='3d'
• Building coordinate grids with np.meshgrid
• Drawing surfaces with ax.plot_surface
• Scattering points in 3D with ax.scatter
• Drawing 3D line plots with ax.plot
• Labeling the x, y, and z axes

1 Creating a 3D Axes

A 3D plot lives on a special Axes. You ask for one by passing projection='3d' when you add the subplot. That single argument unlocks methods like scatter and plot in three dimensions.

📊 What can a 3D Axes draw?

• Scatter clouds of points in space
• Surfaces — a landscape of z heights over an x-y grid
• 3D line plots and curves
• Wireframes and bar charts in 3D

Start with a simple 3D scatter:

import numpy as np
import matplotlib.pyplot as plt

# Some random points in 3D
xs = np.random.rand(30)
ys = np.random.rand(30)
zs = np.random.rand(30)

fig = plt.figure()
# The key: ask for a 3D projection
ax = fig.add_subplot(111, projection="3d")

ax.scatter(xs, ys, zs)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
plt.show()

What you'll see: a cube-shaped plot box with 30 dots floating inside it at random x, y, and z positions. Because it's a 3D Axes, you can rotate it with the mouse in an interactive window, and all three axes are clearly labeled X, Y, and Z.

2 Surface Plots with meshgrid

A surface is a height (z) at every point on an x-y grid. np.meshgrid builds that grid for you, and ax.plot_surface drapes a colored sheet over it.

import numpy as np
import matplotlib.pyplot as plt

# 1D ranges for the two horizontal axes
x = np.linspace(-5, 5, 50)
y = np.linspace(-5, 5, 50)

# Expand into a full 2D coordinate grid
X, Y = np.meshgrid(x, y)

# A height for every (X, Y) point
Z = np.sin(np.sqrt(X**2 + Y**2))

fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")

ax.plot_surface(X, Y, Z, cmap="viridis")

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
plt.show()

What you'll see: a smooth, rippling surface that looks like waves spreading out from the center — the classic "sombrero" shape of sin(sqrt(x² + y²)). The viridis colormap shades high peaks yellow and low troughs purple.

3 3D Line Plots

The same ax.plot you know from 2D works in 3D too — just pass a third list of z values. A parametric curve like a helix is a perfect example.

import numpy as np
import matplotlib.pyplot as plt

# A spiral climbing upward
t = np.linspace(0, 20, 200)
x = np.cos(t)
y = np.sin(t)
z = t

fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")

# plot() takes x, y AND z in 3D
ax.plot(x, y, z)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z (height)")
ax.set_title("A 3D Helix")
plt.show()

What you'll see: a spring-like spiral coiling upward. As t grows, x and y trace a circle while z keeps rising, so the line winds around and climbs — exactly like a stretched-out helix.

🎯 Your Turn: Fill in the Blanks

Replace each ___ to create a 3D Axes and scatter points.

import numpy as np
import matplotlib.pyplot as plt

xs = np.random.rand(20)
ys = np.random.rand(20)
zs = np.random.rand(20)

fig = plt.figure()
# Ask for a 3D projection
ax = fig.add_subplot(111, projection="___")

# Scatter the points in 3D
ax.___(xs, ys, zs)

ax.set_zlabel("Z")
plt.show()

📋 Common Errors & Quick Reference

❌ AttributeError: 'Axes' object has no attribute 'set_zlabel'

You forgot projection='3d'. A plain 2D Axes has no z-axis, so 3D methods don't exist on it.

❌ plot_surface gives a shape error

X, Y, and Z must all be 2D and the same shape. Build X and Y with np.meshgrid, then compute Z from them.

🏆 Mini Challenge: A Paraboloid Surface

Build a bowl-shaped surface from z = x² + y² and label all three axes.

import numpy as np
import matplotlib.pyplot as plt

x = np.linspace(-3, 3, 40)
y = np.linspace(-3, 3, 40)
X, Y = np.meshgrid(x, y)

# 1. Height is a bowl shape
Z = X**2 + Y**2

# 2. Make a 3D Axes
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")

# 3. Draw the surface and label it
ax.plot_surface(X, Y, Z, cmap="plasma")
ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
plt.show()

import matplotlib.pyplot as plt

xs = [1, 2, 3, 4]
ys = [4, 3, 2, 1]
zs = [1, 4, 9, 16]

fig = plt.figure()
ax = fig.add_subplot(projection="3d")
ax.scatter(xs, ys, zs)
ax.set_xlabel("x")
ax.set_ylabel("y")
ax.set_zlabel("z")

print("Projection:", ax.name)
print("Axis labels:", ax.get_xlabel(), ax.get_ylabel(), ax.get_zlabel())
print("Points:", len(zs))
print("Highest z:", max(zs), "at (x, y) =", (xs[zs.index(max(zs))], ys[zs.index(max(zs))]))
print("Collections on the axes:", len(ax.collections))
print("Three axes, not two:", hasattr(ax, "get_zlim"))

# ✅ Expected output:
# Projection: 3d
# Axis labels: x y z
# Points: 4
# Highest z: 16 at (x, y) = (4, 1)
# Collections on the axes: 1
# Three axes, not two: True

❓ Frequently Asked Questions

Lesson complete — your charts now have depth!

You created 3D Axes, scattered points in space, drew surfaces from a meshgrid, plotted a helix, and labeled all three axes.

Practice quiz

Continue this course

• Previous: Heatmaps & imshow
• Next: Customizing with rcParams — Set global defaults for fonts, colors, and sizes with rcParams
• Quick reference: Matplotlib cheat sheet

Frequently asked questions

How do I make a 3D plot in Matplotlib?

Create an Axes with a 3D projection: ax = fig.add_subplot(111, projection='3d'). That ax then has 3D methods like plot_surface, scatter, and plot for drawing in x, y, and z.

Do I need to import Axes3D?

In modern Matplotlib, projection='3d' is enough on its own. Some older code still writes 'from mpl_toolkits.mplot3d import Axes3D' — it does no harm but is no longer required.

What is np.meshgrid used for in 3D plots?

Surface plots need a full grid of x and y coordinates. np.meshgrid takes two 1D arrays and expands them into 2D coordinate grids X and Y, so you can compute a Z value at every point.

How do I label the three axes?

Use ax.set_xlabel(), ax.set_ylabel(), and the 3D-only ax.set_zlabel() to name the x, y, and z axes respectively.