The ultimate training toolkit for finetuning diffusion models
This report summarizes scanner evidence for ostris/ai-toolkit, a Python project with 12,265 GitHub stars. SynthScan examined 223,411 lines of code across 769 source files, recording 2548 pattern matches distributed across 22 syntactic categories. The legacy adjusted pattern signal is 17.4 (Elevated pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 1161 finding(s).
Are imports declared and package/API names verified? 728 finding(s).
Are wrappers, guards, and repeated structures necessary? 1272 finding(s).
Which changes need focused human review? 2548 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 2548 distinct pattern matches across 22 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | run_mac.zsh | 11 | # ── Banner ───────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | run_mac.zsh | 25 | # ── 1. Ensure uv (prebuilt static binary, kept inside the repo) ───── | COMMENT |
| MEDIUM | run_mac.zsh | 54 | # ── 3. Sync the environment and start the UI ──────────────────────── | COMMENT |
| MEDIUM | run_linux.sh | 12 | # ── Banner ───────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | run_linux.sh | 26 | # ── 1. Ensure uv (prebuilt static binary, kept inside the repo) ───── | COMMENT |
| MEDIUM | run_linux.sh | 55 | # ── 3. Sync the environment and start the UI ──────────────────────── | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 31 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 33 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 78 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 80 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 436 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 438 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 526 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 528 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 989 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 991 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1190 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1192 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1238 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1240 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1358 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1360 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1482 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/ace_step/src/model.py | 1484 | # ═══════════════════════════════════════════════════════════════════════════════ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 249 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 251 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 339 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 341 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 407 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 409 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 587 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 590 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 706 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/audio_models/yue2/yue2_model.py | 708 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 782 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 784 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …s_built_in/audio_models/yue2/src/sheetsage_generate.py | 139 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 179 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 181 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 320 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 323 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 404 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 406 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 521 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 523 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 563 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 565 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 640 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 642 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 704 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 706 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/diffusion_models/krea2/krea2.py | 702 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/diffusion_models/krea2/krea2.py | 704 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/diffusion_models/krea2/krea2.py | 706 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | extensions_built_in/diffusion_models/krea2/krea2.py | 708 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/diffusion_models/krea2/krea2.py | 208 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/diffusion_models/krea2/krea2.py | 210 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/diffusion_models/krea2/krea2.py | 407 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/diffusion_models/krea2/krea2.py | 409 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | extensions_built_in/diffusion_models/krea2/krea2.py | 470 | # ------------------------------------------------------------------ | COMMENT |
| 239 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | extensions_built_in/audio_models/base_audio_model.py | 1 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 6 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 6 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 15 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 15 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 22 | CODE | |
| LOW | extensions_built_in/audio_models/ace_step/__init__.py | 1 | CODE | |
| LOW | extensions_built_in/audio_models/ace_step/__init__.py | 1 | CODE | |
| LOW | …ensions_built_in/audio_models/ace_step/src/pipeline.py | 4 | CODE | |
| LOW | …ensions_built_in/audio_models/ace_step/src/pipeline.py | 5 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/__init__.py | 1 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/src/tokenizer.py | 12 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/src/pipeline.py | 24 | CODE | |
| LOW | …ns_built_in/audio_models/yue2/src/shims/pretty_midi.py | 10 | CODE | |
| LOW | …built_in/audio_models/yue2/src/shims/mir_eval_chord.py | 28 | CODE | |
| LOW | extensions_built_in/concept_replacer/ConceptReplacer.py | 3 | CODE | |
| LOW | extensions_built_in/concept_replacer/ConceptReplacer.py | 4 | CODE | |
| LOW | extensions_built_in/concept_replacer/ConceptReplacer.py | 4 | CODE | |
| LOW | extensions_built_in/concept_replacer/ConceptReplacer.py | 5 | CODE | |
| LOW | …nsions_built_in/concept_slider/ConceptSliderTrainer.py | 7 | CODE | |
| LOW | …ons_built_in/diffusion_models/qwen_image_2/__init__.py | 1 | CODE | |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 68 | CODE | |
| LOW | …built_in/diffusion_models/qwen_image_2/src/pipeline.py | 18 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 33 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 50 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 52 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 52 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 62 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/__init__.py | 1 | CODE | |
| LOW | …nsions_built_in/diffusion_models/krea2/src/pipeline.py | 14 | CODE | |
| LOW | extensions_built_in/diffusion_models/chroma/__init__.py | 1 | CODE | |
| LOW | extensions_built_in/diffusion_models/chroma/__init__.py | 2 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 6 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 17 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 18 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 19 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 20 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 21 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 26 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 6 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 16 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 16 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 17 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 18 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 19 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 20 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 20 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 26 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 9 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 9 | CODE | |
| LOW | …nsions_built_in/diffusion_models/ideogram4/__init__.py | 1 | CODE | |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 5 | CODE | |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 24 | CODE | |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 27 | CODE | |
| LOW | …built_in/diffusion_models/ideogram4/src/latent_norm.py | 1 | CODE | |
| LOW | …built_in/diffusion_models/ideogram4/src/transformer.py | 9 | CODE | |
| LOW | …ns_built_in/diffusion_models/ideogram4/src/pipeline.py | 8 | CODE | |
| LOW | extensions_built_in/diffusion_models/ltx2/__init__.py | 1 | CODE | |
| LOW | extensions_built_in/diffusion_models/ltx2/__init__.py | 1 | CODE | |
| LOW | extensions_built_in/diffusion_models/ltx2/__init__.py | 1 | CODE | |
| 664 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 225 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW | extensions_built_in/audio_models/ace_step/src/model.py | 959 | def enable_gradient_checkpointing(self): | CODE |
| LOW | extensions_built_in/audio_models/yue2/yue2_model.py | 318 | def get_text_embedding_space_version(self): | CODE |
| LOW⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 330 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 785 | def convert_lora_weights_before_save(self, state_dict): | CODE |
| LOW⚡ | extensions_built_in/audio_models/yue2/yue2_model.py | 795 | def convert_lora_weights_before_load(self, state_dict): | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 334 | def get_transformer_block_names(cls): | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 338 | def get_quantization_exclude_modules(cls): | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 349 | def enable_gradient_checkpointing(self): | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 353 | def disable_gradient_checkpointing(self): | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 405 | def _nar_lora_from_safetensors(path: str) -> dict: | CODE |
| LOW | …ns_built_in/audio_models/yue2/src/sheetsage_decoder.py | 126 | def gradient_checkpointing_disable(self): | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 423 | def text_embedding_uses_target_size(self) -> bool: | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 427 | def get_text_embedding_space_version(self) -> str: | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 441 | def _normalize_control_images(self, control_images, batch_size: int) -> List[List]: | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 721 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 724 | def get_quantization_exclude_modules(self): | CODE |
| LOW | …t_in/diffusion_models/qwen_image_2/src/text_encoder.py | 39 | def convert_state_dict_on_load(cls, state_dict): | CODE |
| LOW | …t_in/diffusion_models/qwen_image_2/src/text_encoder.py | 52 | def convert_state_dict_on_save(cls, state_dict): | CODE |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 929 | def get_transformer_block_names(cls): | STRING |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 933 | def get_quantization_exclude_modules(cls): | STRING |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 952 | def convert_state_dict_on_load(cls, state_dict): | STRING |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 1202 | def convert_state_dict_on_load(cls, state_dict): | STRING |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 354 | def get_quantization_exclude_modules(self): | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 536 | def _batch_ref_latents_from_batch( | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 728 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/src/mmdit.py | 419 | def get_offload_ignore_modules(self): | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/src/mmdit.py | 493 | def enable_gradient_checkpointing(self): | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/src/mmdit.py | 496 | def disable_gradient_checkpointing(self): | CODE |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 68 | def get_transformer_block_names(self): | CODE |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 45 | def get_transformer_block_names(self): | CODE |
| LOW | …tensions_built_in/diffusion_models/chroma/src/model.py | 58 | def modify_mask_to_attend_padding(mask, max_seq_length, num_extra_padding=8): | CODE |
| LOW | …tensions_built_in/diffusion_models/chroma/src/model.py | 97 | def aitk_config_from_state_dict(cls, state_dict): | CODE |
| LOW | …tensions_built_in/diffusion_models/chroma/src/model.py | 122 | def get_transformer_block_names(cls): | CODE |
| LOW | …tensions_built_in/diffusion_models/chroma/src/model.py | 209 | def enable_gradient_checkpointing(self, enable: bool = True): | CODE |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 72 | def modify_mask_to_attend_padding(mask, max_seq_length, num_extra_padding=8): | CODE |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 111 | def aitk_config_from_state_dict(cls, state_dict): | CODE |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 136 | def get_transformer_block_names(cls): | CODE |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 257 | def enable_gradient_checkpointing(self, enable: bool = True): | CODE |
| LOW | …ensions_built_in/diffusion_models/chroma/src/layers.py | 463 | def _modulation_shift_scale_fn(x, scale, shift): | CODE |
| LOW | …ensions_built_in/diffusion_models/chroma/src/layers.py | 521 | def modulation_shift_scale_fn(self, x, scale, shift): | CODE |
| LOW | …ensions_built_in/diffusion_models/chroma/src/layers.py | 649 | def modulation_shift_scale_fn(self, x, scale, shift): | CODE |
| LOW | …ensions_built_in/diffusion_models/chroma/src/layers.py | 704 | def modulation_shift_scale_fn(self, x, scale, shift): | CODE |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 116 | def _load_component_state_dict(base: str, subfolder: str, basename: str) -> dict: | CODE |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 202 | def text_embedding_space_version(self): | CODE |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 576 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW | …built_in/diffusion_models/ideogram4/src/transformer.py | 364 | def get_transformer_block_names(cls): | CODE |
| LOW | …built_in/diffusion_models/ideogram4/src/transformer.py | 367 | def get_offload_ignore_modules(self): | CODE |
| LOW⚡ | …built_in/diffusion_models/ideogram4/src/transformer.py | 425 | def enable_gradient_checkpointing(self): | CODE |
| LOW⚡ | …built_in/diffusion_models/ideogram4/src/transformer.py | 428 | def disable_gradient_checkpointing(self): | CODE |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 133 | def update_state_dict_inplace( | CODE |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 206 | def split_transformer_and_connector_state_dict( | CODE |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 919 | def get_model_state_dict_from_combined_ckpt( | CODE |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 1068 | def convert_lora_original_to_diffusers( | CODE |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 1102 | def convert_lora_diffusers_to_original( | CODE |
| LOW⚡ | extensions_built_in/diffusion_models/ltx2/ltx2.py | 1147 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| LOW⚡ | extensions_built_in/diffusion_models/ltx2/ltx2.py | 1152 | def convert_lora_weights_before_save(self, state_dict): | CODE |
| LOW⚡ | extensions_built_in/diffusion_models/ltx2/ltx2.py | 1156 | def convert_lora_weights_before_load(self, state_dict): | CODE |
| LOW | …ilt_in/diffusion_models/prx_pixel_t2i/prx_pixel_t2i.py | 277 | def get_latent_noise_from_latents(self, latents: torch.Tensor, noise_offset=0.0): | CODE |
| LOW | …ilt_in/diffusion_models/prx_pixel_t2i/prx_pixel_t2i.py | 315 | def get_transformer_block_names(self) -> Optional[List[str]]: | CODE |
| 417 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …sions_built_in/audio_models/yue2/src/shims/__init__.py | 14 | except ImportError: | CODE |
| HIGH | …ensions_built_in/diffusion_models/omnigen2/__init__.py | 217 | except Exception as e: | CODE |
| HIGH | …ons_built_in/diffusion_models/minimax_h3/minimax_h3.py | 1386 | except Exception: | CODE |
| HIGH | …_in/diffusion_models/minimax_h3/src/ref_video_cache.py | 271 | except Exception: | CODE |
| HIGH | …elines/hidream_image/pipeline_hidream_image_editing.py | 1091 | except Exception as e: | CODE |
| HIGH | …elines/hidream_image/pipeline_hidream_image_editing.py | 1101 | except Exception as e: | CODE |
| HIGH | …t_in/diffusion_models/ming_image/src/bailing_moe_v2.py | 444 | except RuntimeError: | CODE |
| HIGH | extensions_built_in/dataset_tools/tools/image_tools.py | 35 | except Exception as e: | CODE |
| HIGH | extensions_built_in/inference_engine/engine.py | 249 | except Exception: | CODE |
| HIGH | extensions_built_in/inference_engine/engine.py | 648 | except ValueError: | CODE |
| HIGH | extensions_built_in/inference_engine/InferenceEngine.py | 144 | except FileNotFoundError: | CODE |
| HIGH | extensions_built_in/inference_engine/InferenceEngine.py | 161 | except Exception: | CODE |
| HIGH | extensions_built_in/inference_engine/InferenceEngine.py | 90 | except Exception: | CODE |
| HIGH | extensions_built_in/inference_engine/InferenceEngine.py | 149 | except Exception: | CODE |
| HIGH | extensions_built_in/sd_trainer/SDTrainer.py | 1051 | except: | CODE |
| HIGH | extensions_built_in/sd_trainer/SDTrainer.py | 446 | except: | CODE |
| HIGH | extensions_built_in/sd_trainer/DiffusionTrainer.py | 310 | except Exception as e: | CODE |
| HIGH | extensions_built_in/sd_trainer/UITrainer.py | 212 | except Exception as e: | CODE |
| HIGH | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 273 | except RuntimeError: | CODE |
| HIGH | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 722 | except Exception: | CODE |
| HIGH | extensions_built_in/captioner/BaseCaptioner.py | 510 | except Exception as e: | CODE |
| HIGH | extensions_built_in/captioner/Qwen25OmniCaptioner.py | 223 | except Exception: | CODE |
| HIGH | testing/stage_profiler.py | 58 | except Exception: | CODE |
| HIGH | testing/test_ltx_dataloader.py | 219 | except Exception: | CODE |
| HIGH | testing/test_ltx_dataloader.py | 224 | except Exception: | CODE |
| HIGH | manager/env.py | 653 | except (OSError, subprocess.TimeoutExpired): | CODE |
| HIGH | manager/spec.py | 283 | except (TypeError, ValueError): | CODE |
| HIGH | manager/doctor.py | 159 | except OSError: | CODE |
| HIGH | manager/doctor.py | 38 | except RuntimeError: | CODE |
| HIGH | toolkit/ui_job_status.py | 125 | except Exception: | CODE |
| HIGH | toolkit/stable_diffusion_model.py | 2403 | except Exception as e: | CODE |
| HIGH | toolkit/stable_diffusion_model.py | 1313 | except: | CODE |
| HIGH | toolkit/stable_diffusion_model.py | 2241 | except Exception as e: | CODE |
| HIGH | toolkit/kohya_model_util.py | 1399 | except: | CODE |
| HIGH | toolkit/cuda_malloc.py | 82 | except: | CODE |
| HIGH | toolkit/lora_special.py | 625 | except: | CODE |
| HIGH | toolkit/accelerator.py | 18 | except Exception as e: | CODE |
| HIGH | toolkit/train_tools.py | 637 | except Exception as e: | CODE |
| HIGH | toolkit/models/base_model.py | 506 | except: | CODE |
| HIGH | toolkit/models/base_model.py | 1028 | except Exception as e: | CODE |
| HIGH | toolkit/models/base_model.py | 1103 | except Exception as e: | CODE |
| HIGH | toolkit/models/v2/_mixin.py | 573 | except OSError: | CODE |
| HIGH | toolkit/models/v2/pool.py | 38 | except Exception: | CODE |
| HIGH | toolkit/models/v2/pool.py | 188 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 475 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 157 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 327 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 72 | except (TypeError, RuntimeError): | CODE |
| HIGH | toolkit/memory_management/manager.py | 366 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 467 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager.py | 380 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager_modules.py | 170 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager_modules.py | 193 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager_modules.py | 247 | except RuntimeError: | CODE |
| HIGH | toolkit/memory_management/manager_modules.py | 226 | except Exception: | CODE |
| HIGH | toolkit/memory_management/manager_modules.py | 776 | except RuntimeError: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …nsions_built_in/diffusion_models/krea2/src/pipeline.py | 0 | lightweight sampler used by ai-toolkit's preview generation. | STRING |
| HIGH | …ns_built_in/diffusion_models/ideogram4/src/pipeline.py | 0 | lightweight sampler used by ai-toolkit's preview generation. | STRING |
| HIGH | …ons_built_in/diffusion_models/mageflow/src/pipeline.py | 0 | lightweight sampler used by ai-toolkit's preview generation. | STRING |
| HIGH | …_built_in/diffusion_models/boogu_image/src/pipeline.py | 0 | lightweight sampler used by ai-toolkit's preview generation. | STRING |
| HIGH | …sions_built_in/diffusion_models/iris3b/src/pipeline.py | 0 | lightweight sampler used by ai-toolkit's preview generation. | STRING |
| HIGH | …ensions_built_in/diffusion_models/chroma/src/layers.py | 0 | create sinusoidal timestep embeddings. :param t: a 1-d tensor of n indices, one per batch element. these may be fraction | STRING |
| HIGH | extensions_built_in/diffusion_models/flux2/src/model.py | 0 | create sinusoidal timestep embeddings. :param t: a 1-d tensor of n indices, one per batch element. these may be fraction | STRING |
| HIGH | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 0 | create sinusoidal timestep embeddings. :param t: a 1-d tensor of n indices, one per batch element. these may be fraction | STRING |
| HIGH | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 0 | omnigen2 attention processor module copyright 2025 baai, the omnigen2 team and the huggingface team. all rights reserved | STRING |
| HIGH | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 0 | omnigen2 attention processor module copyright 2025 baai, the omnigen2 team and the huggingface team. all rights reserved | STRING |
| HIGH | …sion_models/omnigen2/src/models/attention_processor.py | 0 | omnigen2 attention processor module copyright 2025 baai, the omnigen2 team and the huggingface team. all rights reserved | STRING |
| HIGH | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 0 | calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. handles custom t | STRING |
| HIGH | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 0 | calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. handles custom t | STRING |
| HIGH | …elines/hidream_image/pipeline_hidream_image_editing.py | 0 | calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. handles custom t | STRING |
| HIGH | …/src/pipelines/hidream_image/pipeline_hidream_image.py | 0 | calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. handles custom t | STRING |
| HIGH | …src/schedulers/scheduling_flow_match_euler_discrete.py | 0 | output class for the scheduler's `step` function output. args: prev_sample (`torch.floattensor` of shape `(batch_size, n | STRING |
| HIGH | …sion_models/hidream/src/schedulers/flash_flow_match.py | 0 | output class for the scheduler's `step` function output. args: prev_sample (`torch.floattensor` of shape `(batch_size, n | STRING |
| HIGH | toolkit/samplers/mean_flow_scheduler.py | 0 | output class for the scheduler's `step` function output. args: prev_sample (`torch.floattensor` of shape `(batch_size, n | STRING |
| HIGH | …src/schedulers/scheduling_flow_match_euler_discrete.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …sion_models/hidream/src/schedulers/flash_flow_match.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …sion_models/hidream/src/schedulers/fm_solvers_unipc.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …src/schedulers/scheduling_flow_match_euler_discrete.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …sion_models/hidream/src/schedulers/flash_flow_match.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …sion_models/hidream/src/schedulers/fm_solvers_unipc.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …src/schedulers/scheduling_flow_match_euler_discrete.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | …sion_models/hidream/src/schedulers/flash_flow_match.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | …sion_models/hidream/src/schedulers/fm_solvers_unipc.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 0 | "dynamic thresholding: at each sampling step we set s to a certain percentile absolute pixel value in xt0 (the predictio | STRING |
| HIGH | …sion_models/hidream/src/schedulers/fm_solvers_unipc.py | 0 | "dynamic thresholding: at each sampling step we set s to a certain percentile absolute pixel value in xt0 (the predictio | STRING |
| HIGH | toolkit/samplers/custom_lcm_scheduler.py | 0 | "dynamic thresholding: at each sampling step we set s to a certain percentile absolute pixel value in xt0 (the predictio | STRING |
| HIGH | …usion_models/hidream/src/models/attention_processor.py | 0 | attention processor used typically in processing the sd3-like self-attention projections. | STRING |
| HIGH | toolkit/ip_adapter.py | 0 | attention processor used typically in processing the sd3-like self-attention projections. | STRING |
| HIGH | toolkit/models/flux_sage_attn.py | 0 | attention processor used typically in processing the sd3-like self-attention projections. | STRING |
| HIGH | toolkit/models/vd_adapter.py | 0 | attention processor used typically in processing the sd3-like self-attention projections. | STRING |
| HIGH | extensions_built_in/sd_trainer/DiffusionTrainer.py | 0 | start a daemon thread that periodically checks should_stop() and terminates the process immediately when triggered. | STRING |
| HIGH | extensions_built_in/sd_trainer/UITrainer.py | 0 | start a daemon thread that periodically checks should_stop() and terminates the process immediately when triggered. | STRING |
| HIGH | extensions_built_in/captioner/BaseCaptioner.py | 0 | start a daemon thread that periodically checks should_stop() and terminates the process immediately when triggered. | STRING |
| HIGH | extensions_built_in/sd_trainer/DiffusionTrainer.py | 0 | helper method to run an async coroutine and track the task. | STRING |
| HIGH | extensions_built_in/sd_trainer/UITrainer.py | 0 | helper method to run an async coroutine and track the task. | STRING |
| HIGH | extensions_built_in/captioner/BaseCaptioner.py | 0 | helper method to run an async coroutine and track the task. | STRING |
| HIGH | extensions_built_in/sd_trainer/DiffusionTrainer.py | 0 | create a new connection for each operation to avoid locking. | STRING |
| HIGH | extensions_built_in/sd_trainer/UITrainer.py | 0 | create a new connection for each operation to avoid locking. | STRING |
| HIGH | extensions_built_in/captioner/BaseCaptioner.py | 0 | create a new connection for each operation to avoid locking. | STRING |
| HIGH | toolkit/reference_adapter.py | 0 | attention processor for custom te for pytorch 2.0. args: hidden_size (`int`): the hidden size of the attention layer. cr | STRING |
| HIGH | toolkit/util/ip_adapter_utils.py | 0 | attention processor for custom te for pytorch 2.0. args: hidden_size (`int`): the hidden size of the attention layer. cr | STRING |
| HIGH | toolkit/models/te_adapter.py | 0 | attention processor for custom te for pytorch 2.0. args: hidden_size (`int`): the hidden size of the attention layer. cr | STRING |
| HIGH | toolkit/util/ip_adapter_utils.py | 0 | processor for implementing scaled dot-product attention (enabled by default if you're using pytorch 2.0). | STRING |
| HIGH | toolkit/models/single_value_adapter.py | 0 | processor for implementing scaled dot-product attention (enabled by default if you're using pytorch 2.0). | STRING |
| HIGH | toolkit/models/vd_adapter.py | 0 | processor for implementing scaled dot-product attention (enabled by default if you're using pytorch 2.0). | STRING |
| HIGH | toolkit/optimizers/automagic.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r | STRING |
| HIGH | toolkit/optimizers/adafactor.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r | STRING |
| HIGH | toolkit/optimizers/prodigy_8bit.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r | STRING |
| HIGH | toolkit/optimizers/adam8bit.py | 0 | performs a single optimization step. arguments: closure (callable, optional): a closure that reevaluates the model and r | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | flux_train_ui.py | 25 | CODE | |
| LOW | ui_scripts/upsample_ideogram4_caption.py | 132 | CODE | |
| LOW | ui_scripts/merge_loras.py | 36 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/yue2_model.py | 630 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 358 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/src/tokenizer.py | 79 | CODE | |
| LOW | …s_built_in/audio_models/yue2/src/sheetsage_generate.py | 36 | CODE | |
| LOW | …ns_built_in/audio_models/yue2/src/shims/pretty_midi.py | 35 | CODE | |
| LOW | …built_in/audio_models/yue2/src/shims/mir_eval_chord.py | 69 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 284 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 323 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 630 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 1641 | CODE | |
| LOW | extensions_built_in/diffusion_models/chroma/pipeline.py | 148 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/chroma_model.py | 101 | CODE | |
| LOW | …lt_in/diffusion_models/chroma/chroma_radiance_model.py | 78 | CODE | |
| LOW | …tensions_built_in/diffusion_models/chroma/src/model.py | 97 | CODE | |
| LOW | …sions_built_in/diffusion_models/chroma/src/radiance.py | 111 | CODE | |
| LOW | …ensions_built_in/diffusion_models/chroma/src/layers.py | 92 | CODE | |
| LOW | …ns_built_in/diffusion_models/ideogram4/src/pipeline.py | 312 | CODE | |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 919 | CODE | |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 974 | CODE | |
| LOW | …_in/diffusion_models/ltx2/convert_ltx2_to_diffusers.py | 1020 | CODE | |
| LOW | extensions_built_in/diffusion_models/ltx2/ltx2.py | 821 | CODE | |
| LOW | …ensions_built_in/diffusion_models/flux2/flux2_model.py | 286 | CODE | |
| LOW | extensions_built_in/diffusion_models/anima/anima.py | 203 | CODE | |
| LOW | extensions_built_in/diffusion_models/iris3b/src/dit.py | 604 | CODE | |
| LOW | …in/diffusion_models/qwen_image/qwen_image_edit_plus.py | 251 | CODE | |
| LOW | …in/diffusion_models/qwen_image/qwen_image_pipelines.py | 28 | CODE | |
| LOW | …usion_models/omnigen2/src/pipelines/image_processor.py | 135 | CODE | |
| LOW | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 341 | CODE | |
| LOW | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 709 | CODE | |
| LOW | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 346 | CODE | |
| LOW | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 607 | CODE | |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 205 | CODE | |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 493 | CODE | |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 595 | CODE | |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 664 | CODE | |
| LOW | …lt_in/diffusion_models/boogu_image/boogu_image_edit.py | 188 | CODE | |
| LOW | …iffusion_models/boogu_image/src/attention_processor.py | 264 | CODE | |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 964 | CODE | |
| LOW | …ons_built_in/diffusion_models/minimax_h3/minimax_h3.py | 538 | CODE | |
| LOW | …ons_built_in/diffusion_models/minimax_h3/minimax_h3.py | 772 | CODE | |
| LOW | …ilt_in/diffusion_models/minimax_h3/src/text_encoder.py | 117 | CODE | |
| LOW | …s_built_in/diffusion_models/minimax_h3/src/pipeline.py | 65 | CODE | |
| LOW | …nsions_built_in/diffusion_models/minimax_h3/src/vae.py | 715 | CODE | |
| LOW | …built_in/diffusion_models/z_image/z_image_l2p_model.py | 291 | CODE | |
| LOW | …ions_built_in/diffusion_models/wan22/wan22_5b_model.py | 250 | CODE | |
| LOW | …ions_built_in/diffusion_models/wan22/wan22_pipeline.py | 46 | CODE | |
| LOW | …ons_built_in/diffusion_models/wan22/wan22_14b_model.py | 466 | CODE | |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 569 | CODE | |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 773 | CODE | |
| LOW | …/src/pipelines/hidream_image/pipeline_hidream_image.py | 516 | CODE | |
| LOW | …in/diffusion_models/hidream/src/hidream_o1/pipeline.py | 32 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 1260 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 2345 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 2368 | CODE | |
| LOW | …ilt_in/diffusion_models/ming_image/src/text_encoder.py | 219 | CODE | |
| LOW | …ilt_in/diffusion_models/ming_image/src/text_encoder.py | 296 | CODE | |
| LOW | …ilt_in/diffusion_models/ming_image/src/text_encoder.py | 367 | CODE | |
| 246 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | run.py | 127 | except Exception as e: | CODE |
| LOW⚡ | run.py | 132 | except Exception as e2: | CODE |
| LOW⚡ | run.py | 140 | except Exception as e2: | CODE |
| LOW | run_modal.py | 135 | except Exception as e: | CODE |
| MEDIUM | run_modal.py | 136 | print(f"Error running job: {e}") | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/model.py | 433 | except Exception: | CODE |
| LOW | extensions_built_in/audio_models/yue2/src/pipeline.py | 201 | except Exception as e: # capture unsupported for these ops/kernels: eager path | CODE |
| LOW | …built_in/diffusion_models/qwen_image_2/qwen_image_2.py | 279 | except Exception as e: | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 292 | except Exception as e: | CODE |
| LOW | …sions_built_in/diffusion_models/ideogram4/ideogram4.py | 302 | except Exception as e: | CODE |
| LOW | …ensions_built_in/diffusion_models/omnigen2/__init__.py | 217 | except Exception as e: | CODE |
| LOW | …ons_built_in/diffusion_models/minimax_h3/minimax_h3.py | 317 | except Exception as e: | CODE |
| LOW | …ons_built_in/diffusion_models/minimax_h3/minimax_h3.py | 1386 | except Exception: | CODE |
| LOW | …_in/diffusion_models/minimax_h3/src/ref_video_cache.py | 271 | except Exception: | CODE |
| LOW | …ilt_in/diffusion_models/minimax_h3/src/text_encoder.py | 50 | except Exception: | CODE |
| MEDIUM | …ilt_in/diffusion_models/minimax_h3/src/text_encoder.py | 44 | def video_has_audio(path) -> bool: | CODE |
| LOW | …nsions_built_in/diffusion_models/minimax_h3/src/vsa.py | 53 | except Exception: | CODE |
| LOW | extensions_built_in/diffusion_models/z_image/z_image.py | 95 | except Exception as e: | CODE |
| LOW | …_built_in/diffusion_models/hidream/hidream_o1_model.py | 179 | except Exception as e: | CODE |
| LOW⚡ | …elines/hidream_image/pipeline_hidream_image_editing.py | 1091 | except Exception as e: | STRING |
| LOW⚡ | …elines/hidream_image/pipeline_hidream_image_editing.py | 1101 | except Exception as e: | STRING |
| LOW | …ons_built_in/diffusion_models/ming_image/ming_image.py | 224 | except Exception as e: | CODE |
| LOW | extensions_built_in/dataset_tools/SuperTagger.py | 184 | except Exception: | CODE |
| LOW | extensions_built_in/dataset_tools/SyncFromCollection.py | 83 | except Exception as e: | CODE |
| LOW | extensions_built_in/dataset_tools/SyncFromCollection.py | 110 | except Exception as e: | CODE |
| LOW | extensions_built_in/dataset_tools/tools/image_tools.py | 35 | except Exception as e: | CODE |
| LOW | extensions_built_in/inference_engine/engine.py | 214 | except Exception: | CODE |
| LOW | extensions_built_in/inference_engine/engine.py | 249 | except Exception: | CODE |
| LOW | extensions_built_in/inference_engine/engine.py | 272 | except Exception as e: | CODE |
| LOW | extensions_built_in/inference_engine/engine.py | 688 | except Exception as e: | CODE |
| LOW | extensions_built_in/inference_engine/engine.py | 735 | except Exception as e: | CODE |
| LOW | extensions_built_in/inference_engine/InferenceEngine.py | 90 | except Exception: | CODE |
| LOW | extensions_built_in/inference_engine/InferenceEngine.py | 149 | except Exception: | CODE |
| LOW | extensions_built_in/inference_engine/InferenceEngine.py | 161 | except Exception: | CODE |
| LOW | extensions_built_in/sd_trainer/SDTrainer.py | 150 | except Exception: | CODE |
| LOW | extensions_built_in/sd_trainer/SDTrainer.py | 1016 | except Exception as e: | CODE |
| LOW | extensions_built_in/sd_trainer/DiffusionTrainer.py | 84 | except Exception: | CODE |
| LOW | extensions_built_in/sd_trainer/DiffusionTrainer.py | 310 | except Exception as e: | CODE |
| LOW | extensions_built_in/sd_trainer/DiffusionTrainer.py | 334 | except Exception as db_err: | CODE |
| LOW | extensions_built_in/sd_trainer/UITrainer.py | 76 | except Exception: | CODE |
| LOW | extensions_built_in/sd_trainer/UITrainer.py | 212 | except Exception as e: | CODE |
| LOW | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 722 | except Exception: | CODE |
| LOW | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 881 | except Exception as e: | CODE |
| LOW | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 914 | except Exception as e: | CODE |
| MEDIUM | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 915 | print(f"Error preprocessing {path}: {e}") | CODE |
| LOW | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 963 | except Exception as e: | CODE |
| MEDIUM | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 964 | print(f"Error processing {file_path}: {e}") | CODE |
| LOW | extensions_built_in/captioner/Qwen3VLCaptioner.py | 156 | except Exception as e: | CODE |
| MEDIUM | extensions_built_in/captioner/Qwen3VLCaptioner.py | 157 | print(f"Error processing {file_path}: {e}") | CODE |
| LOW⚡ | extensions_built_in/captioner/AceStepCaptioner.py | 631 | except Exception as e: | CODE |
| MEDIUM⚡ | extensions_built_in/captioner/AceStepCaptioner.py | 632 | print(f"Error preprocessing {path}: {e}") | CODE |
| LOW⚡ | extensions_built_in/captioner/AceStepCaptioner.py | 637 | except Exception as e: | CODE |
| MEDIUM⚡ | extensions_built_in/captioner/AceStepCaptioner.py | 638 | print(f"Error processing {path}: {e}") | CODE |
| LOW | extensions_built_in/captioner/AceStepCaptioner.py | 649 | except Exception as e: | CODE |
| MEDIUM | extensions_built_in/captioner/AceStepCaptioner.py | 650 | print(f"Error processing {file_path}: {e}") | CODE |
| LOW | extensions_built_in/captioner/BaseCaptioner.py | 136 | except Exception as e: | CODE |
| MEDIUM | extensions_built_in/captioner/BaseCaptioner.py | 137 | print(f"Error captioning file {file_path}: {e}") | CODE |
| LOW | extensions_built_in/captioner/BaseCaptioner.py | 258 | except Exception as e: | CODE |
| LOW | extensions_built_in/captioner/BaseCaptioner.py | 329 | except Exception: | CODE |
| LOW | extensions_built_in/captioner/BaseCaptioner.py | 496 | except Exception as db_err: | CODE |
| 164 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | flux_train_ui.py | 358 | "Include sample prompts to test out your trained model. Don't forget to include your trigger word/senten | STRING |
| MEDIUM | config/examples/generate.example.yaml | 22 | # here ate the flags you can use for prompts. Always start with | COMMENT |
| MEDIUM | config/examples/train_lora_flex_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flex_24gb.yaml | 82 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flex_24gb.yaml | 99 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_flux_kontext_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_kontext_24gb.yaml | 74 | # and then you can use this model. | COMMENT |
| MEDIUM | config/examples/train_lora_flux_kontext_24gb.yaml | 86 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_kontext_24gb.yaml | 104 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_sd35_large_24gb.yaml | 15 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_sd35_large_24gb.yaml | 78 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_sd35_large_24gb.yaml | 95 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | …fig/examples/train_lora_qwen_image_edit_2509_32gb.yaml | 103 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_flex.yaml | 15 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_flex.yaml | 88 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_flex.yaml | 105 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_chroma_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_chroma_24gb.yaml | 85 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_chroma_24gb.yaml | 102 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_hidream_48.yaml | 22 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_hidream_48.yaml | 93 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_hidream_48.yaml | 110 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_14b_24gb.yaml | 18 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_14b_24gb.yaml | 91 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_14b_24gb.yaml | 99 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_omnigen2_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_omnigen2_24gb.yaml | 75 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_omnigen2_24gb.yaml | 92 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_lumina.yaml | 15 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_lumina.yaml | 77 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_lumina.yaml | 94 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_wan22_14b_24gb.yaml | 92 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_wan22_14b_24gb.yaml | 109 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_1b_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_1b_24gb.yaml | 80 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_wan21_1b_24gb.yaml | 88 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_flex_redux.yaml | 48 | # for normal redux, we are just recreating the same image, so you can use the same folder path above | COMMENT |
| MEDIUM | config/examples/train_flex_redux.yaml | 54 | # this is what I used for the 24GB card, but feel free to adjust | COMMENT |
| MEDIUM | config/examples/train_flex_redux.yaml | 110 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_flux_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_24gb.yaml | 77 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_24gb.yaml | 94 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_qwen_image_edit_32gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_qwen_image_edit_32gb.yaml | 100 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_flux_schnell_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_schnell_24gb.yaml | 79 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flux_schnell_24gb.yaml | 96 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_lumina.yaml | 15 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_lumina.yaml | 80 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_full_fine_tune_lumina.yaml | 97 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_flex2_24gb.yaml | 18 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flex2_24gb.yaml | 136 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_flex2_24gb.yaml | 139 | # you can use a single inpaint or single control image on your samples. | COMMENT |
| MEDIUM | config/examples/train_lora_flex2_24gb.yaml | 163 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/train_lora_qwen_image_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_qwen_image_24gb.yaml | 76 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/train_lora_qwen_image_24gb.yaml | 93 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| MEDIUM | config/examples/modal/modal_train_lora_flux_24gb.yaml | 14 | # alternatively, in your captions you can add [trigger] and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/modal/modal_train_lora_flux_24gb.yaml | 77 | # you can add [trigger] to the prompts here and it will be replaced with the trigger word | COMMENT |
| MEDIUM | config/examples/modal/modal_train_lora_flux_24gb.yaml | 94 | # you can add any additional meta info here. [name] is replaced with config name at top | COMMENT |
| 4 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 269 | CODE | |
| LOW | …ns_built_in/audio_models/ace_step/ace_step_15_model.py | 320 | CODE | |
| LOW | extensions_built_in/audio_models/ace_step/src/model.py | 1378 | CODE | |
| LOW | extensions_built_in/audio_models/ace_step/src/model.py | 891 | CODE | |
| LOW | …ensions_built_in/audio_models/ace_step/src/pipeline.py | 75 | CODE | |
| LOW | extensions_built_in/audio_models/yue2/src/pipeline.py | 232 | CODE | |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 716 | CODE | |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 975 | CODE | |
| LOW | …built_in/diffusion_models/qwen_image_2/src/pipeline.py | 350 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 630 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 929 | CODE | |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 1218 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/src/mmdit.py | 269 | CODE | |
| LOW | extensions_built_in/diffusion_models/krea2/src/mmdit.py | 611 | CODE | |
| LOW | extensions_built_in/diffusion_models/chroma/pipeline.py | 148 | CODE | |
| LOW | …/diffusion_models/prx_pixel_t2i/src/transformer_prx.py | 717 | CODE | |
| LOW | …nsions_built_in/diffusion_models/flux2/src/pipeline.py | 286 | CODE | |
| LOW | …ensions_built_in/diffusion_models/f_light/src/model.py | 317 | CODE | |
| LOW | …ions_built_in/diffusion_models/f_light/src/pipeline.py | 160 | CODE | |
| LOW | …sions_built_in/diffusion_models/iris3b/src/pipeline.py | 81 | CODE | |
| LOW | extensions_built_in/diffusion_models/iris3b/src/dit.py | 639 | CODE | |
| LOW | extensions_built_in/diffusion_models/iris3b/src/dit.py | 704 | CODE | |
| LOW | …in/diffusion_models/qwen_image/qwen_image_pipelines.py | 28 | CODE | |
| LOW | …_built_in/diffusion_models/mageflow/src/transformer.py | 260 | CODE | |
| LOW | …tensions_built_in/diffusion_models/mageflow/src/vae.py | 437 | CODE | |
| LOW | …tensions_built_in/diffusion_models/mageflow/src/vae.py | 499 | CODE | |
| LOW | …usion_models/omnigen2/src/pipelines/image_processor.py | 49 | CODE | |
| LOW | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 341 | CODE | |
| LOW | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 494 | CODE | |
| LOW | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 638 | CODE | |
| LOW | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 346 | CODE | |
| LOW | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 462 | CODE | |
| LOW | …nigen2/src/models/transformers/transformer_omnigen2.py | 225 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 54 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 114 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 298 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 611 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 1012 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 1052 | CODE | |
| LOW | …diffusion_models/omnigen2/src/ops/triton/layer_norm.py | 756 | CODE | |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 205 | CODE | |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 518 | CODE | |
| LOW | …_built_in/diffusion_models/boogu_image/src/pipeline.py | 130 | CODE | |
| LOW | …s_built_in/diffusion_models/minimax_h3/src/pipeline.py | 65 | CODE | |
| LOW | …nsions_built_in/diffusion_models/minimax_h3/src/vae.py | 457 | CODE | |
| LOW | …ions_built_in/diffusion_models/wan22/wan22_pipeline.py | 46 | CODE | |
| LOW | …n/diffusion_models/zeta_chroma/zeta_chroma_pipeline.py | 60 | CODE | |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 341 | CODE | |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 569 | CODE | |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 773 | CODE | |
| LOW | …/src/pipelines/hidream_image/pipeline_hidream_image.py | 291 | CODE | |
| LOW | …/src/pipelines/hidream_image/pipeline_hidream_image.py | 516 | CODE | |
| LOW | …lt_in/diffusion_models/hidream/src/models/attention.py | 9 | CODE | |
| LOW | …m/src/models/transformers/transformer_hidream_image.py | 245 | CODE | |
| LOW | …in/diffusion_models/hidream/src/hidream_o1/pipeline.py | 290 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 950 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 1876 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 2158 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 2245 | CODE | |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 2381 | CODE | |
| 76 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | spark_requirements.txt | 1 | # NVIDIA RTX Spark — native Windows-on-ARM (win_arm64) requirement set. | COMMENT |
| LOW | ui/src/utils/terminalEmulator.ts | 1 | // Minimal terminal emulator for rendering job logs the way a real terminal | COMMENT |
| LOW | config/examples/generate.example.yaml | 21 | COMMENT | |
| LOW | config/examples/train_lora_flux_kontext_24gb.yaml | 21 | dtype: float16 # precision to save | COMMENT |
| LOW | config/examples/train_lora_hidream_48.yaml | 1 | # HiDream training is still highly experimental. The settings here will take ~35.2GB of vram to train. | COMMENT |
| LOW | config/examples/train_lora_wan21_14b_24gb.yaml | 1 | # IMPORTANT: The Wan2.1 14B model is huge. This config should work on 24GB GPUs. It cannot | COMMENT |
| LOW | config/examples/train_slider.example.yml | 1 | --- | COMMENT |
| LOW | config/examples/train_slider.example.yml | 41 | # for sliders we are adjusting representation of the concept (unet), | COMMENT |
| LOW | config/examples/train_slider.example.yml | 81 | save_every: 50 # save every this many steps | COMMENT |
| LOW | config/examples/train_slider.example.yml | 141 | # slider training config, best for last | COMMENT |
| LOW | config/examples/train_slider.example.yml | 161 | # for example, if we are adjusting the representation of a person, we would use "person" | COMMENT |
| LOW | config/examples/train_slider.example.yml | 181 | # max 76 tokens for now | COMMENT |
| LOW | config/examples/train_slider.example.yml | 201 | # regardless if they are smiling or not, however, the closer the concept is to the target_class | COMMENT |
| LOW | config/examples/train_slider.example.yml | 221 | # It is saved in the model so be aware of that. The software will include this | COMMENT |
| LOW | config/examples/train_lora_flex2_24gb.yaml | 41 | # this will automatically generate the controls for you before training. The current script is not | COMMENT |
| LOW | …ns_built_in/concept_replacer/config/train.example.yaml | 81 | # You can put any information you want here, and it will be saved in the model. | COMMENT |
| LOW | …lt_in/diffusion_models/qwen_image_2/src/transformer.py | 1 | # Copyright 2026 Qwen-Image Team, The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 1 | # Copyright 2026 The Qwen Team and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | extensions_built_in/diffusion_models/krea2/krea2.py | 181 | # image sequence at t=0 (ComfyUI Kontext "index_timestep_zero"). Runs in | COMMENT |
| LOW | …sions_built_in/diffusion_models/chroma/src/__init__.py | 1 | # This is taken and slightly modified from https://github.com/lodestone-rock/flow/tree/master/src/models/chroma | COMMENT |
| LOW | …/diffusion_models/prx_pixel_t2i/src/transformer_prx.py | 1 | # PRXTransformer2DModel -- vendored, self-contained copy of the Photoroom PRX | COMMENT |
| LOW | …usion_models/omnigen2/src/pipelines/image_processor.py | 1 | # Copyright 2024 The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …in/diffusion_models/omnigen2/src/utils/import_utils.py | 1 | # Copyright 2024 The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …_in/diffusion_models/omnigen2/src/models/embeddings.py | 1 | # Copyright 2024 The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …dels/omnigen2/src/models/transformers/block_lumina2.py | 1 | COMMENT | |
| LOW | …src/schedulers/scheduling_flow_match_euler_discrete.py | 1 | # Copyright 2024 Stability AI, Katherine Crowson and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …igen2/src/schedulers/scheduling_dpmsolver_multistep.py | 1 | # Copyright 2024 TSAIL Team and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …minimax_h3/src/vsa_kernels/block_sparse_attn_triton.py | 21 | # ──────────────────────────── SPARSE ADDITION BEGIN ─────────────────────────── | COMMENT |
| LOW | …ns_built_in/diffusion_models/hidream/src/models/moe.py | 81 | # scores_for_seq_aux = scores_for_aux.view(bsz, seq_len, -1) | COMMENT |
| LOW | …ns_built_in/diffusion_models/hidream/src/models/moe.py | 121 | x = x.view(-1, x.shape[-1]) | COMMENT |
| LOW | …models/hidream/src/hidream_o1/qwen3_vl_transformers.py | 1 | # Copyright 2025 The Qwen Team and The HuggingFace Inc. team. All rights reserved. | COMMENT |
| LOW | …sion_models/hidream/src/schedulers/flash_flow_match.py | 1 | # Copyright 2024 Stability AI, Katherine Crowson and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …uilt_in/diffusion_models/ming_image/src/transformer.py | 1 | # Copyright 2025 Alibaba Z-Image Team and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …ilt_in/diffusion_models/example_model/example_model.py | 81 | # Class names of modules whose Linear layers get LoRA'd. Matched against | COMMENT |
| LOW | …ns_built_in/diffusion_models/example_model/__init__.py | 1 | # This is a documentation-only TEMPLATE model. Start with README.md in this | COMMENT |
| LOW | …s_built_in/diffusion_models/ernie_image/transformer.py | 1 | # Copyright 2025 Baidu ERNIE-Image Team and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | extensions_built_in/dataset_tools/tools/fuyu_utils.py | 61 | # # autoregressively generate text | COMMENT |
| LOW | extensions_built_in/dataset_tools/tools/sync_tools.py | 201 | # input_path = img_path | COMMENT |
| LOW | extensions_built_in/dataset_tools/tools/sync_tools.py | 221 | # img = cv2.imread(img_path) | COMMENT |
| LOW | extensions_built_in/dataset_tools/tools/sync_tools.py | 241 | # os.makedirs(os.path.dirname(img_path), exist_ok=True) | COMMENT |
| LOW | extensions_built_in/dataset_tools/tools/sync_tools.py | 261 | # # update eq captions | COMMENT |
| LOW | …_built_in/advanced_generator/config/train.example.yaml | 81 | # You can put any information you want here, and it will be saved in the model. | COMMENT |
| LOW | extensions_built_in/sd_trainer/SDTrainer.py | 2321 | print_acc("loss is nan") | COMMENT |
| LOW | …tensions_built_in/sd_trainer/config/train.example.yaml | 81 | # You can put any information you want here, and it will be saved in the model. | COMMENT |
| LOW | extensions_built_in/captioner/Qwen3OmniCaptioner.py | 321 | top_k = top_k_index.shape[-1] | COMMENT |
| LOW | …ltimate_slider_trainer/UltimateSliderTrainerProcess.py | 361 | # with torch.set_grad_enabled(self.train_config.train_text_encoder): | COMMENT |
| LOW | extensions/example/ExampleMergeModels.py | 41 | # checking requirements. This is called before the run() function | COMMENT |
| LOW | testing/test_bucket_dataloader.py | 61 | dataset_config = DatasetConfig( | COMMENT |
| LOW | testing/test_bucket_dataloader.py | 101 | COMMENT | |
| LOW | scripts/convert_diffusers_to_comfy_transformer_only.py | 1 | ####################################################### | COMMENT |
| LOW | scripts/test_quantizations.py | 41 | import torch # noqa: E402 | COMMENT |
| LOW | scripts/convert_diffusers_to_comfy.py | 1 | ####################################################### | COMMENT |
| LOW | jobs/process/TrainVAEProcess.py | 861 | COMMENT | |
| LOW | jobs/process/BaseSDTrainProcess.py | 1821 | torch.backends.cuda.enable_mem_efficient_sdp(True) | COMMENT |
| LOW | jobs/process/TrainSliderProcess.py | 461 | current_timestep, | COMMENT |
| LOW | jobs/process/TrainSliderProcess.py | 481 | # prompt_pair.positive_target, # negative prompt | COMMENT |
| LOW | toolkit/config_modules.py | 1061 | # threads that decode/resize the images of one bucketed batch concurrently inside a | COMMENT |
| LOW | toolkit/config_modules.py | 1441 | # --d Specify the seed for the generated image. | COMMENT |
| LOW | toolkit/data_loader.py | 781 | # boot new worker processes every epoch, which can take longer than the epoch | COMMENT |
| LOW | toolkit/inversion_utils.py | 241 | return noisy_sample, zs, xts | COMMENT |
| 33 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | …ions_built_in/diffusion_models/qwen_image_2/src/vae.py | 543 | # Create the components | STRING |
| MEDIUM | …ensions_built_in/diffusion_models/chroma/src/layers.py | 229 | # Create a 2D meshgrid of coordinates. | COMMENT |
| MEDIUM | …ensions_built_in/diffusion_models/chroma/src/layers.py | 237 | # Create a 1D tensor of frequency values from 0 to max_freqs-1. | COMMENT |
| MEDIUM | extensions_built_in/sd_trainer/DiffusionTrainer.py | 35 | # Create a thread pool for database operations | COMMENT |
| MEDIUM | extensions_built_in/sd_trainer/DiffusionTrainer.py | 96 | # Create a task and track it | COMMENT |
| MEDIUM | extensions_built_in/sd_trainer/UITrainer.py | 29 | # Create a thread pool for database operations | COMMENT |
| MEDIUM | extensions_built_in/sd_trainer/UITrainer.py | 88 | # Create a task and track it | COMMENT |
| MEDIUM | extensions_built_in/captioner/BaseCaptioner.py | 83 | # Create a thread pool for database operations | COMMENT |
| MEDIUM | extensions_built_in/captioner/BaseCaptioner.py | 341 | # Create a task and track it | COMMENT |
| MEDIUM | testing/shrink_pixart_sm.py | 43 | # Create a new state dict for the reduced model | COMMENT |
| MEDIUM | testing/shrink_pixart_sm3.py | 35 | # Create a new state dict for the reduced model | COMMENT |
| MEDIUM | testing/shrink_pixart_sm2.py | 45 | # Create a new state dict for the reduced model | COMMENT |
| MEDIUM | scripts/convert_diffusers_to_comfy_transformer_only.py | 361 | # Define the float8 range | COMMENT |
| MEDIUM | scripts/convert_diffusers_to_comfy.py | 355 | # Define the float8 range | COMMENT |
| MEDIUM | toolkit/style.py | 34 | # Define the separate loss function | COMMENT |
| MEDIUM | toolkit/dataloader_mixins.py | 92 | # Define the normalization transform | COMMENT |
| MEDIUM | toolkit/dataloader_mixins.py | 1253 | # Create a background with the specified transparent color | COMMENT |
| MEDIUM | toolkit/util/mask.py | 69 | # Create a low-frequency noise field first (for smooth organic shapes) | COMMENT |
| MEDIUM | toolkit/util/shuffle.py | 38 | # Create a slice object to shuffle along the specified axis | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …igen2/src/pipelines/omnigen2/pipeline_omnigen2_chat.py | 286 | Get prompt embeddings from the Qwen2 text encoder. Args: prompt: The prompt or list of pro | STRING |
| HIGH | …s/omnigen2/src/pipelines/omnigen2/pipeline_omnigen2.py | 274 | Get prompt embeddings from the Qwen2 text encoder. Args: prompt: The prompt or list of pro | STRING |
| HIGH | …elines/hidream_image/pipeline_hidream_image_editing.py | 809 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| HIGH | extensions_built_in/flex2/pipeline.py | 52 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| HIGH | toolkit/train_pipelines.py | 45 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| HIGH | toolkit/pipelines.py | 50 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| HIGH | toolkit/pipelines.py | 590 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| HIGH | toolkit/pipelines.py | 1198 | Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | config/examples/train_slider.example.yml | 18 | # just use one for now | COMMENT |
| MEDIUM | config/examples/train_slider.example.yml | 188 | # this will make the LoRA more robust. You probably want this on unless prompt order | COMMENT |
| MEDIUM | config/examples/train_lora_flex2_24gb.yaml | 121 | # if you are not training with it. Controls are a little more robust and can be left out, | COMMENT |
| LOW | extensions_built_in/diffusion_models/ltx2/ltx2.py | 93 | image, # will contain a dict that can be dumped ditectly into encode_video, just add output_path to it. | CODE |
| MEDIUM | …uilt_in/diffusion_models/prx_pixel_t2i/src/pipeline.py | 53 | # called by the sampler harness (inside a try/except, so optional) | COMMENT |
| LOW | …m/src/models/transformers/transformer_hidream_image.py | 330 | # the implementation on hidream during train was wrong, just use the inference one. | COMMENT |
| MEDIUM | …uilt_in/diffusion_models/example_model/src/pipeline.py | 79 | # called by the sampler harness (inside a try/except, so optional); | COMMENT |
| LOW | jobs/process/BaseExtractProcess.py | 60 | # we will just add process it to the end of the filename if there is more than one process | COMMENT |
| LOW | toolkit/stable_diffusion_model.py | 1797 | # todo, should we just use the Text encoder for the refiner? Fine tuned versions will differ | COMMENT |
| LOW | toolkit/stable_diffusion_model.py | 2018 | # if we only have 1 timestep, we can just use the same timestep for all | COMMENT |
| LOW | toolkit/stable_diffusion_model.py | 2113 | # just use the first second text encoder | COMMENT |
| LOW | toolkit/stable_diffusion_model.py | 2128 | # just use the first second text encoder | COMMENT |
| LOW | toolkit/control_generator.py | 122 | # just pass the PIL image straight through. | COMMENT |
| LOW | toolkit/models/base_model.py | 954 | # if we only have 1 timestep, we can just use the same timestep for all | COMMENT |
| LOW | toolkit/models/te_adapter.py | 399 | # just use aura pile | STRING |
| LOW | toolkit/optimizers/optimizer_utils.py | 197 | # If no dtype specified, just pass through to parent | COMMENT |
| MEDIUM | toolkit/optimizers/automagicEXPERIMENT.py | 394 | # unchanged keeps every property that made v3 robust. A window of H | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | run_modal.py | 80 | # Check if we have DEBUG_TOOLKIT in env | COMMENT |
| LOW | run_modal.py | 82 | # Set torch to trace mode | COMMENT |
| LOW | …ions_built_in/diffusion_models/f_light/src/pipeline.py | 224 | # Check if guidance is needed | COMMENT |
| LOW⚡ | …elines/hidream_image/pipeline_hidream_image_editing.py | 1096 | # Check if we need to switch from editing stage to refining stage | STRING |
| LOW | scripts/update_sponsors.py | 274 | # Check if required environment variables are set | COMMENT |
| LOW | toolkit/timer.py | 38 | # Check if this timer's buffer exceeds max_buffer and remove the oldest if it does | COMMENT |
| LOW⚡ | toolkit/extension.py | 42 | # Check if the value is a list | COMMENT |
| LOW⚡ | toolkit/extension.py | 44 | # Iterate over the list and add the classes to the main list | COMMENT |
| LOW | toolkit/dequantize.py | 69 | # Check if the parameter exists | COMMENT |
| LOW | toolkit/train_tools.py | 226 | if not chunk.eq(chunk[0, 0]).all(): # Check if all elements in the chunk are the same as the first element | CODE |
| LOW | toolkit/data_transfer_object/data_loader.py | 136 | # Check if video opened successfully | COMMENT |
| LOW | toolkit/util/get_model.py | 35 | # Check if the value is a list | COMMENT |
| LOW | toolkit/util/get_model.py | 37 | # Iterate over the list and add the classes to the main list | COMMENT |
| LOW | toolkit/optimizers/automagic.py | 185 | # Check if exp_avg_sq_row and exp_avg_sq_col exist for factored case | COMMENT |
| LOW | toolkit/optimizers/automagic.py | 193 | # Check if exp_avg_sq exists for non-factored case | COMMENT |
| LOW | toolkit/optimizers/automagic.py | 323 | # Check if lr_mask exists in the state before trying to access it | COMMENT |
| LOW | toolkit/optimizers/automagic.py | 337 | # Check if state_dict has the expected structure | COMMENT |
| LOW | toolkit/optimizers/automagic.py | 339 | # Check if at least one state entry has an lr_mask, which is specific to Automagic | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ons_built_in/diffusion_models/qwen_image_2/__init__.py | 3 | __all__ = ["QwenImage2Model"] | CODE |
| LOW | extensions_built_in/diffusion_models/krea2/__init__.py | 3 | __all__ = ["Krea2Model"] | CODE |
| LOW⚡ | …built_in/diffusion_models/ideogram4/src/transformer.py | 431 | def set_attention_backend(self, backend: str) -> None: | CODE |
| LOW | …ns_built_in/diffusion_models/prx_pixel_t2i/__init__.py | 3 | __all__ = ["PRXPixelT2IModel"] | CODE |
| LOW | …ions_built_in/diffusion_models/f_light/src/__init__.py | 5 | __all__ = ["FLitePipeline", "FLitePipelineOutput", "APGConfig", "DiT"] | CODE |
| LOW | …ions_built_in/diffusion_models/f_light/src/pipeline.py | 19 | logger = logging.getLogger(__name__) | CODE |
| LOW | extensions_built_in/diffusion_models/iris3b/__init__.py | 3 | __all__ = ["Iris3BModel"] | CODE |
| LOW | …ensions_built_in/diffusion_models/mageflow/src/attn.py | 73 | __all__ = ["flash_attn_varlen_func"] | CODE |
| LOW | …on_models/omnigen2/src/models/transformers/__init__.py | 3 | __all__ = ["OmniGen2Transformer2DModel"] | CODE |
| LOW | …ions_built_in/diffusion_models/boogu_image/__init__.py | 4 | __all__ = ["BooguImageModel", "BooguImageEditModel"] | CODE |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 699 | def set_attention_backend(self, backend: str) -> None: | CODE |
| LOW | …elines/hidream_image/pipeline_hidream_image_editing.py | 34 | logger = logging.getLogger(__name__) | CODE |
| LOW | …sions_built_in/diffusion_models/ming_image/__init__.py | 3 | __all__ = ["MingImageModel"] | CODE |
| LOW | …ns_built_in/diffusion_models/example_model/__init__.py | 12 | __all__ = ["ExampleModel"] | CODE |
| LOW | toolkit/network_mixins.py | 182 | def _set_runtime_scale(self: Module, value) -> None: | CODE |
| LOW | toolkit/kohya_lora.py | 81 | def _set_runtime_scale(self, value) -> None: | CODE |
| LOW | toolkit/optimizers/automagic2.py | 103 | def _update_param(self, p: torch.Tensor, group: dict) -> None: | CODE |
| LOW | toolkit/optimizers/automagic3.py | 413 | def _update_param(self, p: torch.Tensor, group: dict) -> None: | CODE |
| LOW | toolkit/audio/moss_music/configuration.py | 135 | __all__ = ["MossMusicEncoderConfig", "MossMusicConfig"] | CODE |
| LOW | toolkit/audio/moss_music/__init__.py | 27 | __all__ = [ | CODE |
| LOW | toolkit/audio/moss_music/processing.py | 250 | __all__ = ["MelConfig", "MossMusicProcessor"] | CODE |
| LOW | toolkit/audio/moss_music/modeling.py | 564 | __all__ = [ | CODE |
| LOW | toolkit/audio/melbandroformer/__init__.py | 4 | __all__ = [ | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | ui/src/hooks/useJobLossLog.tsx | 43 | // Step 1: get key list (we can do this by calling endpoint once; it returns keys) | COMMENT |
| LOW | ui/src/hooks/useJobLossLog.tsx | 54 | // Step 2: fetch each loss key incrementally (since_step per key if polling) | COMMENT |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 347 | # Step 1: modulation for both streams. | COMMENT |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 364 | # Step 2: joint attention on [instruct + img]. | COMMENT |
| LOW⚡ | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 391 | # Step 3: image self-attention. | COMMENT |
| LOW⚡ | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 399 | # Step 4: residual updates. | COMMENT |
| LOW⚡ | toolkit/basic.py | 50 | # Step 1: Calculate mean and variance of content features | COMMENT |
| LOW⚡ | toolkit/basic.py | 54 | # Step 2: Calculate mean and variance of style features | COMMENT |
| LOW⚡ | toolkit/basic.py | 58 | # Step 3: Normalize content features | COMMENT |
| LOW⚡ | toolkit/basic.py | 62 | # Step 4: Scale and shift normalized content with style's statistics | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | ui/src/hooks/useJobLossLog.tsx | 43 | // Step 1: get key list (we can do this by calling endpoint once; it returns keys) | COMMENT |
| LOW | ui/src/hooks/useJobLossLog.tsx | 54 | // Step 2: fetch each loss key incrementally (since_step per key if polling) | COMMENT |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 347 | # Step 1: modulation for both streams. | COMMENT |
| LOW | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 364 | # Step 2: joint attention on [instruct + img]. | COMMENT |
| LOW⚡ | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 391 | # Step 3: image self-attention. | COMMENT |
| LOW⚡ | …ilt_in/diffusion_models/boogu_image/src/transformer.py | 399 | # Step 4: residual updates. | COMMENT |
| LOW⚡ | toolkit/basic.py | 50 | # Step 1: Calculate mean and variance of content features | COMMENT |
| LOW⚡ | toolkit/basic.py | 54 | # Step 2: Calculate mean and variance of style features | COMMENT |
| LOW⚡ | toolkit/basic.py | 58 | # Step 3: Normalize content features | COMMENT |
| LOW⚡ | toolkit/basic.py | 62 | # Step 4: Scale and shift normalized content with style's statistics | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | ui_scripts/upsample_ideogram4_caption.py | 295 | '({"index": i, "caption": {...}|null}) instead of a single final list.' | CODE |
| HIGH | extensions_built_in/audio_models/yue2/yue2_model.py | 507 | """Next-token CE over ``ids`` (and KL(base || lora) when ``ar_kl_weight`` > 0), computed in | STRING |
| HIGH | manager/env.py | 633 | 'cmd /s /c "call "%s" >nul && set"' % vcvars, | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | toolkit/ip_adapter.py | 1152 | yield from self.image_encoder.vision_model.embeddings.position_embedding.parameters(recurse) | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | config/examples/train_slider.example.yml | 166 | # you can leave it blank to affect everything. In this example, we are adjusting | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | extensions_built_in/inference_engine/PLANNING.md | 126 | engine.py # Engine: request queue, holder cache, pool orchestration | CODE |
| MEDIUM | toolkit/models/lokr.py | 351 | # org_forward here does raw CUDA stream/event orchestration and | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | toolkit/unloader.py | 22 | CODE |