Alexander Eichhorn ba2475c3f0 fix(z-image): improve device/dtype compatibility and error handling ... Add robust device capability detection for bfloat16, replacing hardcoded dtype with runtime checks that fallback to float16/float32 on unsupported hardware. This prevents runtime failures on GPUs and CPUs without bfloat16. Key changes: - Add TorchDevice.choose_bfloat16_safe_dtype() helper for safe dtype selection - Fix LoRA device mismatch in layer_patcher.py (add device= to .to() call) - Replace all assert statements with descriptive exceptions (TypeError/ValueError) - Add hidden_states bounds check and apply_chat_template fallback in text encoder - Add GGUF QKV tensor validation (divisible by 3 check) - Fix CPU noise generation to use float32 for compatibility - Remove verbose debug logging from LoRA conversion utils		2025-12-09 07:37:06 +01:00
..
app	fix(z-image): improve device/dtype compatibility and error handling	2025-12-09 07:37:06 +01:00
assets
backend	fix(z-image): improve device/dtype compatibility and error handling	2025-12-09 07:37:06 +01:00
configs
frontend	feat(z-image): add metadata recall for VAE and Qwen3 encoder	2025-12-09 07:12:36 +01:00
invocation_api	feat(mm): more exports in invocation api	2025-10-16 08:08:44 +11:00
version	chore: bump version to v6.9.0	2025-10-17 12:13:01 +11:00
__init__.py