Environment: Ubuntu (kernel 7.0), RTX 5090 (Blackwell sm_120), NVIDIA driver w/ CUDA 13.3 userland, Modly AppImage v0.4.x, embedded Python 3.11.9, data dir relocated to a large secondary disk. Found while driving Modly headlessly (REST + MCP) for a game-asset pipeline.
If an enable_texture generation fails during its lazy texgen setup (e.g. missing xatlas), the extension worker is left with self._model = None. Every subsequent generation on that worker then raises:
TypeError: 'NoneType' object is not callable
…until the worker process is manually killed.
Suggested fix: guard generate() with a model-loaded check that triggers a reload, or recycle the worker when setup fails.
Environment: Ubuntu (kernel 7.0), RTX 5090 (Blackwell sm_120), NVIDIA driver w/ CUDA 13.3 userland, Modly AppImage v0.4.x, embedded Python 3.11.9, data dir relocated to a large secondary disk. Found while driving Modly headlessly (REST + MCP) for a game-asset pipeline.
If an
enable_texturegeneration fails during its lazy texgen setup (e.g. missingxatlas), the extension worker is left withself._model = None. Every subsequent generation on that worker then raises:…until the worker process is manually killed.
Suggested fix: guard
generate()with a model-loaded check that triggers a reload, or recycle the worker when setup fails.