Blender’s Grease Pencil has been slowly turning into a serious 2D animation system rather than a 3D annotation tool. It has strokes as real objects, bone-based deformation, onion skinning and layers. What it hasn’t had is shape keys — and a first-look test published on 27 September shows them arriving.
Why this is the missing piece
Anyone who has rigged a 3D face knows the division of labour. Bones handle rotation and translation — a jaw hinging, a limb swinging. Shape keys (blend shapes, morph targets) handle everything bones can’t express: a smile, a squint, a cheek puffing, a silhouette that needs to be drawn differently rather than transformed.
2D character animation has exactly the same problem, and arguably worse. Rotate a drawn arm with a bone and the shoulder tears, because a 2D drawing has no volume to preserve. Turn a drawn head and it doesn’t turn — it just rotates, like a sticker. Every 2D rigging system hits the point where the correct answer is “at this angle, the shape is simply different,” and without shape keys the only options are drawing a separate frame or accepting the distortion.
With shape keys on Grease Pencil strokes, an animator can rig the motion and author corrective shapes for the poses where the rig breaks down — which is how Moho, Toon Boom and After Effects rigging systems all eventually work.
What “first look” means here
Be clear about the status: this is a test of in-development functionality, not a shipped feature in a stable release. Treat it as a signal about direction rather than something to plan a production around this month.
The test uses a simple, recognisable character rig — which is the right way to evaluate this. A rigging feature demonstrated on an abstract shape tells you nothing; demonstrated on a character with a round head that has to turn, it immediately shows whether the corrective shapes hold up.
What it means for Blender’s position
Blender has quietly become plausible as a 2D-and-3D pipeline in one application, and shape keys on Grease Pencil closes a real gap against dedicated 2D tools.
The practical argument for it isn’t feature parity — Moho and Toon Boom remain more mature at 2D specifically. It’s not leaving the application. A production that does 2D characters over 3D environments, or 2D effects on 3D renders, currently moves data between packages at every stage. Doing the rigging where the lighting and camera already live removes an entire category of round-trip problem.
The counter-argument, worth stating: Grease Pencil’s history is as a 3D-adjacent tool, and 2D animators have well-founded habits built around applications designed for them from the start. A feature arriving is not the same as a workflow being good. This is worth watching, and it isn’t yet a reason to move a production.