Object Call Override Template Create

This is a node from Avatar Graph, Avatech AI's pack for turning a single character portrait into a real-time interactive 2D avatar - animated blinking, lip sync - built on Blender's Python engine (bpy) plus Avatech's own "Shape Flow" runtime for live shape-key control. The pack's README advertises maybe eight nodes as its actual feature set: Create Mesh Layer, Create Shape Flow, Avatar Main Output, and a few others. Object Call Override Template Create isn't one of them. It's from the large, code-generated ObjectCall_* family mirroring Blender's bpy.types.Object API, and this particular method is one of the more obscure corners of that API even by Blender's own standards.

object.override_template_create() sets up an override template on the object - a scaffold used inside Blender's library-override system to define which custom properties or ID-property-driven values are allowed to be overridden on instances of this object elsewhere, rather than creating an actual override itself (that's what Override Create and Override Hierarchy Create do). It's a configuration step deep in Blender's asset-library tooling, not something with an obvious analog in a typical Avatar Graph rigging workflow. Honestly, if you're building an eye-blink-and-lipsync avatar from the pack's own templates, there's essentially no path that leads you here - this reads like a completeness artifact of exposing the whole Object API, not a node anyone's workflow diagram actually calls for.

The schema backs that up - it's the barest node in the whole family: just BPY_OBJ in (optional per the schema, though the node has nothing to act on without one) and BPY_OBJ out, the same object handle passed straight through. No configuration surface at all; the underlying Blender call takes no arguments beyond the object itself.

Installing it is installing Avatar Graph as a whole. Via ComfyUI Manager: search "Avatar Graph" and install. Manually: cd ComfyUI/custom_nodes && git clone https://github.com/avatechai/avatar-graph-comfyui, then cd avatar-graph-comfyui && pip install -r requirements.txt, restart ComfyUI. The one requirement that'll actually get you: bpy needs Python 3.10.x specifically, so the README's push toward a dedicated conda create --name comfyui python=3.10 environment, rather than reusing whatever you've got set up for Flux or SDXL, is worth taking at face value. Restart with --enable-cors-header (add --force-fp16 on macOS) to enable the live avatar preview through Avatech's editor.

Troubleshooting: given how rarely this exact node would ever fire in a real workflow, the problem you're actually likely to hit is upstream of it - bpy failing to import at ComfyUI startup because the interpreter isn't exactly 3.10.x, a genuinely common and recurring failure mode across Blender-Python-based ComfyUI nodes generally, this pack very much included. Skipping --enable-cors-header costs you the live preview panel only - the graph and the .glb/.gltf export from Avatar Main Output still work fine. Beyond that, there's no write-up for this node anywhere in the pack, and honestly if you find yourself needing to reason carefully about override templates, Blender's own manual on the library-override system - not this article - is where that actually gets explained.