UpscalerSeedVR.load_model() rebinds the module-global
generation.generation_step (called by name inside generation_loop) to
the instance's model_step wrapper, keeping the previous value to call
back into. That global was never restored, so the second pass through
load_model() saved the wrapper itself as the "original", making
model_step() call itself -> RecursionError.
The second pass is reached on any model (re)load: with upscaler_unload
enabled (self.model reset to None after each run) every subsequent run
recurses, and switching SeedVR variants (self.model_loaded != model_name)
triggers it even without unload.
Stash the pristine generation_step on the module once and have the
wrapper call that, so repeated loads never wrap the wrapper.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
NetworkModule.__init__ set self.shape only inside 'if hasattr(sd_module, weight)' but then used len(self.shape) unconditionally, raising AttributeError when a LoRA targets a weightless module. Default shape to None and skip the dora_norm_dims computation when absent.
Co-Authored-By: Claude <noreply@anthropic.com>
hr_upscale_to_x is the width dimension and hr_upscale_to_y the height (see processing_class.py); set_resolution assigned them transposed, producing swapped HyperTile geometry on non-square hires passes.
Co-Authored-By: Claude <noreply@anthropic.com>
api_list_models omitted the 'or model_type == all' clause for zimage, so ZImage ControlNets never appeared in the all listing - added it to match every other family. lite_model used pass for the unimplemented root==output branch, which fell through to b = getattr(b, attn_name) with b unbound or stale from a prior iteration (UnboundLocalError or wrong-block patch); use continue to skip it.
Co-Authored-By: Claude <noreply@anthropic.com>
sd_hijack_accelerate.py: bias.to(weight.dtype) discarded its result (Tensor.to is not in-place), so the conv still received the mismatched bias. sd_hijack_te.py: os.environ.get returns a str when MAX_SEQUENCE_LENGTH is set, so max(int, str) raised TypeError (uncaught, before the try); coerce with int(). sd_detect.py: pipeline was referenced in 'if callable(pipeline)' but only assigned when cls is not None, raising UnboundLocalError for a model_index.json without _class_name.
Co-Authored-By: Claude <noreply@anthropic.com>
generate.py: p.ip_adapter_masks was reinitialized inside the per-adapter loop, discarding all but the last adapter's masks; move it beside the other accumulators. process.py: req.params is dict|None, so a null params body crashed .items() in post_preprocess/post_mask. api.py: split(':') without maxsplit broke auth/auth-file entries whose password contains a colon. gallery.py: allowed_paths stored quote(path) but the membership check and path guards use the raw path, causing duplicate accumulation and an ineffective whitelist; also drop the unused FastAPI import (pylint W0611 surfaced when this file is linted).
Co-Authored-By: Claude <noreply@anthropic.com>
make_lora reassigned the 'modules' selection arg to a named_modules() generator, so the subsequent 'te'/'unet' in modules checks tested an exhausted generator and silently skipped TE2 + UNet extraction. Also 'loaded_lora() == ""' never matched a loaded model (returns a list), so the no-LoRA-detected guard never fired.
Co-Authored-By: Claude <noreply@anthropic.com>
The VideoCapture is stored in cap, but the per-frame read and skip/yield guards referenced video, which is set to None at function entry and never reassigned. As a result only the first frame of a video input was processed.
Co-Authored-By: Claude <noreply@anthropic.com>
transformer., bare-diffusers, and lora_transformer_ bases are already in
network-key form for every arch, yet each per-arch resolve_targets repeated the
same passthrough branch for them. Move that into a shared
PASSTHROUGH_PREFIXES_DEFAULT set consulted by resolve_group_targets, leaving each
arch's resolve_targets to only the prefixes it actually rewrites (kohya / BFL).
lycoris_ stays in flux2, the one arch that recognizes it.
Pure refactor: the same keys resolve to the same modules.
Add an end-to-end regression per diffusers arch (chroma, flux2, zimage, ernie)
that loads a lora_transformer_ diffusers-flat state dict and checks the
expected modules bind. The chroma suite also asserts the shared
resolve_group_targets passthrough runs above each arch's own resolve_targets,
which returns nothing for this prefix.
OneTrainer saves LoRAs against the diffusers layout, keying each module as
'lora_transformer_' + the underscore-flattened module path with QKV pre-split.
That is sdnext's own network_layer_mapping namespace, but the native loader did
not list it as a known prefix, so parse_key dropped every key and the network
loaded zero modules ("not loaded").
Add lora_transformer_ to KNOWN_PREFIXES_DEFAULT and resolve it in a shared
resolve_group_targets helper that passes the base through unchanged, with no
rename or chunking. Routing every family loader through the helper gives all
diffusers arches (chroma, flux2, zimage, ernie) OneTrainer support without
per-arch wiring.
Fixes#4877
Disty0
Vladimir Mandic
CalamitousFelicitousness
QualiaRain