Best Infill Pattern for Strength: What the Research and Slicer Docs Actually Show

The search for the best infill pattern for strength produces a lot of conflicting advice online because the question has more than one correct answer. Pattern, density, load direction, and material all interact, and research on this topic — including peer-reviewed tensile testing — shows that the “winner” shifts depending on which variable you change. What follows is a pattern-by-pattern breakdown based on published mechanical testing data and official slicer documentation, not opinion.

Why Pattern Geometry Matters More Than Density Alone

Infill density (the percentage of interior volume that gets filled with plastic) is only half the story. Pattern geometry determines how load transfers through that material, whether stress concentrates at nodes or distributes across surfaces, and how crack propagation behaves.

Raising density from 20% to 40% with a weak pattern will improve strength, but switching from a poorly chosen pattern to a better-matched one at the same density often beats it. The practical implication: before increasing infill density to chase strength, check whether a different pattern at the existing density would serve better. Walls and perimeters almost always contribute more to part strength than infill — bumping from 3 to 5 walls at 20% infill will typically outperform keeping 3 walls and running 40% infill, particularly under bending and tensile loads parallel to the print surface.

Pattern-by-Pattern Breakdown

Gyroid

Gyroid is a triply periodic minimal surface — a continuous, curving structure that repeats in all three axes without crossing itself within a single layer. The PrusaSlicer documentation ↗ describes gyroid as “one of the best infills” precisely because it “provides equal strength in all directions.” This near-isotropic behavior makes it the safest general-purpose strength choice for parts where the load direction is unknown or varies — enclosures, brackets, tool handles, anything subject to torque or multi-axis stress.

Gyroid also prints faster than honeycomb and uses material efficiently. It’s the pattern to reach for when you don’t want to think too hard about part orientation relative to load.

Recommended starting point: 20–25% for general functional parts, 30–35% for higher-stress applications.

Cubic and Adaptive Cubic

Cubic infill creates a 3D grid of rotated cubes that provides decent strength in all three axes. Adaptive cubic adjusts density automatically — denser near walls, sparser in the interior — reducing material use by roughly 25% compared to uniform cubic at the same nominal density. According to Prusa’s documentation, adaptive cubic is a solid choice for large prints where weight reduction matters.

Research published in PMC ↗ testing four PLA patterns at 60% infill density via ASTM D638 dog-bone specimens found adaptive cubic achieved 13.05 ± 0.58 MPa ultimate tensile strength — the lowest of the four patterns tested. This doesn’t mean cubic is useless; it means the geometry doesn’t concentrate load-bearing paths along a single axis, which is a tradeoff for multi-directional capability.

Honeycomb

Honeycomb creates hexagonal columns aligned along the Z-axis. It provides excellent compressive resistance when the load pushes straight down the column axis, which is why it performs well in jigs, press-fit fixtures, and structural blocks oriented with the print’s Z-axis matching the compression direction.

The same PrusaSlicer documentation notes that honeycomb requires approximately 25% more material than gyroid and can take roughly twice as long to print. The PMC12788003 study found honeycomb achieved 18.42 ± 0.77 MPa tensile strength in PLA at 60% infill, second-highest among the four tested patterns, but only when load aligns favorably. If your part will see bending or off-axis forces, honeycomb’s column geometry becomes a liability rather than an asset.

Grid and Rectilinear

Grid and rectilinear infills lay down straight lines — grid does this in both X and Y each layer, while rectilinear alternates direction. They’re the fastest patterns to print and the most material-efficient at a given density.

The same PLA tensile study found grid infill produced the highest ultimate tensile strength at 21.91 ± 0.61 MPa ↗ across all four patterns tested at 60% density — beating honeycomb, rectilinear, and adaptive cubic. The reason is straightforward: continuous, aligned load paths along the loading axis maximize the cross-sectional area carrying tension. When the load direction is predictable and the part is oriented correctly, grid or rectilinear will beat gyroid on a raw tensile number.

The catch: if the load rotates, shifts, or applies bending moments, the aligned geometry becomes a weakness. Grid is the right call for flat brackets, beams, and straps under known unidirectional tension — not for anything that will see complex loading.

Concentric

Concentric infill traces the perimeter shape inward in parallel offsets. It doesn’t look like it would be strong given the gaps between rings, but its geometry interrupts crack propagation more effectively than linear patterns.

Research on ABS specimens (PMC10238722 ↗) found concentric at 80% infill density with 100 µm layer height achieved 38.95 MPa ultimate tensile strength, compared to 17.46 MPa for line and 18.05 MPa for triangle patterns at similar densities — a 123% and 115% improvement respectively. Impact strength gains were even larger. The circular geometry distributes stress around the perimeter rather than concentrating it at line intersections.

Concentric is the right pattern for functional parts that take repeated impact or shock loads: phone mounts, tool grips, protective cases. At densities below 40%, its ring structure has gaps that limit load transfer, so it needs higher fill percentages to realize the strength advantage.

Matching Pattern to Application

Use case	Pattern	Infill %
Unknown load direction / general functional	Gyroid	20–25%
Unidirectional tension (beams, straps)	Grid or rectilinear	25–35%
Compression (jigs, clamps, press-fits)	Honeycomb or cubic	25–35%
Impact / shock loading	Concentric	35–50%
Large prints, weight-sensitive	Adaptive cubic	15–25%

For density above 60%, the marginal strength gain from infill drops sharply — the walls are carrying most of the load by that point. If you need more strength at high density, add walls (perimeters) and top/bottom solid layers instead of pushing infill percentage past 60%.

One more variable worth dialing: layer height. The ABS study (PMC10238722 ↗) found 100 µm layer height outperformed coarser layers for tensile strength. Thinner layers increase interlayer bond surface area and reduce void volume at layer interfaces. For strength-critical prints, running 0.15–0.20 mm layers instead of the default 0.20–0.25 mm is worth the extra print time.

Sources

• PrusaSlicer Infill Patterns — Prusa Knowledge Base ↗ — Official documentation covering gyroid, honeycomb, cubic, adaptive cubic, and rectilinear, including print time and material comparisons.

• Comprehensive Investigation of Infill Geometry Effects on Mechanical Performance of Polymer 3D Printed Components (PMC12788003) ↗ — Peer-reviewed PLA tensile testing of grid, honeycomb, rectilinear, and adaptive cubic at 60% infill density per ASTM D638; grid achieved 21.91 MPa, adaptive cubic 13.05 MPa.

• Combined Effect of Infill Pattern, Density, and Layer Thickness on Mechanical Properties of 3D Printed ABS (PMC10238722) ↗ — ABS study showing concentric infill at 80% density, 100 µm layer height achieved 38.95 MPa tensile strength — 123% above line infill at the same density.