Gravia (v0.8.10)

Gravity becomes both your weapon and your worst enemy in Gravia, a compact physics-action puzzler built around the dangerous idea of carrying a portable singularity. Armed with the Gravia Glove, you create temporary Rifts that drag nearby objects inward, turning quiet test chambers into chaotic scenes of collapsing machinery, flying debris, and improvised launches. The tone is more clever than grim, but there is real tension in every experiment: one badly placed black hole can clear a path, fling you to safety, or pull you straight into disaster. Its minimalist setup keeps the focus firmly on manipulating forces rather than following a story.

Each level asks you to reach an exit by placing Rifts intelligently and reacting to the physical consequences. Loose objects spiral toward the gravitational pull, while heavier or reinforced structures can bend, snap, and become tools of their own when enough force is applied. Players use that destruction to remove barriers, build momentum for huge jumps, reposition crates, and solve environmental problems that rarely have only one obvious answer. Roughly 50 stages provide a campaign that lasts around five hours, while difficult hidden Secrets encourage repeated attempts and demand far more precise control than simply finishing a level.

What makes Gravia memorable is how naturally action and puzzle-solving overlap; launching yourself through a room feels exciting, yet success still depends on understanding angles, timing, mass, and collateral damage. Its simulation-driven design produces the kind of accidental solutions and spectacular failures associated with physics sandboxes, but within tightly constructed puzzle spaces rather than an open world. Time Trial mode adds another layer for players who enjoy optimizing routes and shaving seconds from runs after mastering each chamber. Fans of Portal’s experimental problem-solving, or the destructive improvisation of games like Teardown, find plenty to enjoy here, especially if they appreciate challenging secrets and are willing to learn through explosive trial and error.