Home / Lightricks / LTX-2
AI Slop Code RIsk Monitor

Lightricks/LTX-2

Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.

35.3 High pattern signal View on GitHub

Maintenance and Governance Review

This report summarizes scanner evidence for Lightricks/LTX-2, a Python project with 9,648 GitHub stars. SynthScan examined 93,263 lines of code across 425 source files, recording 1982 pattern matches distributed across 21 syntactic categories. The legacy adjusted pattern signal is 35.3 (High 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.

35.3
Adjusted Signal
35.3
Raw Pattern Score
100%
Time Factor
2026-10-02
Last Push
9.6K
Stars
Python
Language
93.3K
Lines of Code
425
Files
2.0K
Pattern Hits
2026-10-12
Scan Date
0.05
HC Hit Rate

Risk Review · v3

Use each vector to guide manual verification of the underlying findings.

Cognitive Ownership Risk 29.4/100 · MEDIUM

Can maintainers explain intent, assumptions, and trade-offs? 690 finding(s).

Hallucinated Dependency & Supply Chain Risk 26.7/100 · MEDIUM

Are imports declared and package/API names verified? 560 finding(s).

Agent Over-Engineering Ratio 73.0/100 · HIGH

Are wrappers, guards, and repeated structures necessary? 1117 finding(s).

PR Review Velocity Impact 83.2/100 · CRITICAL

Which changes need focused human review? 1982 finding(s).

What These Metrics Mean

Adjusted Signal
Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score
Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor
A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits
Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate
High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files
Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

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.

Severity Breakdown

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.

CRITICAL 6HIGH 14MEDIUM 612LOW 1350

Directory Score Breakdown

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.

Pattern Findings

The scanner identified 1982 distinct pattern matches across 21 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.

Decorative Section Separators602 hits · 1900 pts
SeverityFileLineSnippetContext
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml26# =============================================================================COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml28# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml30# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml327# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml329# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml339# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml341# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml351# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/v2a_lora.yaml353# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml1# =============================================================================COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml3# =============================================================================COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml54# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml56# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml112# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml114# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml143# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml145# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml180# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml182# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml203# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml205# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml216# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml218# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml310# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml312# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml371# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/v2a_lora.yaml373# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml26# =============================================================================COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml28# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml30# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml321# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml323# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml333# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml335# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml345# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml347# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml1# =============================================================================COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml3# =============================================================================COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml54# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml56# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml107# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml109# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml133# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml135# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml170# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml172# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml193# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml195# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml206# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml208# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml304# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml306# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml365# -----------------------------------------------------------------------------COMMENT
MEDIUMpackages/ltx-trainer/configs/audio_extend_lora.yaml367# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml26# =============================================================================COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml28# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml30# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml321# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml323# -----------------------------------------------------------------------------COMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml333# -----------------------------------------------------------------------------COMMENT
542 more matches not shown…
Unused Imports548 hits · 509 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/src/ltx_trainer/gemma_8bit.py21CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py16CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py33CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py33CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py33CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py34CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py35CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py35CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py36CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py37CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py79CODE
LOWpackages/ltx-trainer/src/ltx_trainer/training_state.py1CODE
LOW…rainer/src/ltx_trainer/training_strategies/__init__.py12CODE
LOW…rainer/src/ltx_trainer/training_strategies/__init__.py12CODE
LOW…rainer/src/ltx_trainer/training_strategies/__init__.py12CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/distilled_mgpu.py11CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py5CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/retake.py1CODE
LOW…-pipelines/src/ltx_pipelines/ti2vid_two_stages_mgpu.py10CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py25CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py26CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py27CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py28CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py29CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py30CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py31CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py32CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py33CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py34CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py35CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dubit.py3CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/dfr_pipeline.py3CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dfr_stages.py12CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/hdr_ic_lora.py3CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dfr_mgpu.py11CODE
LOW…pelines/src/ltx_pipelines/ti2vid_two_stages_hq_mgpu.py13CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/multigpu/runner.py7CODE
LOW…ltx-pipelines/src/ltx_pipelines/multigpu/controller.py31CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/multigpu/fleet.py9CODE
LOW…tx-pipelines/src/ltx_pipelines/multigpu/tdp_builder.py5CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py12CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py12CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py12CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py12CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py12CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py19CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py20CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py20CODE
LOW…pipelines/src/ltx_pipelines/multigpu/gemma_builders.py15CODE
LOW…ltx-pipelines/src/ltx_pipelines/multigpu/sp_builder.py6CODE
LOW…pelines/src/ltx_pipelines/multigpu/bp_gemma_builder.py22CODE
LOW…ltx-pipelines/src/ltx_pipelines/multigpu/_broadcast.py7CODE
LOW…tx-pipelines/src/ltx_pipelines/multigpu/nccl_groups.py3CODE
LOW…x-pipelines/src/ltx_pipelines/multigpu/vae_builders.py5CODE
LOW…lines/src/ltx_pipelines/multigpu/delegating_builder.py6CODE
LOW…pipelines/src/ltx_pipelines/multigpu/weight_tracker.py5CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/__init__.py1CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/__init__.py1CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/__init__.py1CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/__init__.py1CODE
488 more matches not shown…
Hyper-Verbose Identifiers320 hits · 314 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/scripts/compute_reference.py169 def media_path_to_reference_path(media_file: Path) -> Path:CODE
LOWpackages/ltx-trainer/scripts/process_captions.py168 def _load_caption_data_from_csv(self) -> dict[str, str]:CODE
LOWpackages/ltx-trainer/scripts/process_captions.py185 def _load_caption_data_from_json(self) -> dict[str, str]:CODE
LOWpackages/ltx-trainer/scripts/process_captions.py206 def _load_caption_data_from_jsonl(self) -> dict[str, str]:CODE
LOWpackages/ltx-trainer/scripts/process_captions.py233def compute_captions_embeddings( # noqa: PLR0913CODE
LOWpackages/ltx-trainer/scripts/process_captions.py376def _build_sharded_dataloader(CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py60 def validate_exactly_one_modality(self) -> "PrefixConditionConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py81 def validate_exactly_one_modality(self) -> "SuffixConditionConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py106 def validate_exactly_one_modality(self) -> "MaskConditionConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py123 def validate_exactly_one_modality(self) -> "ReferenceConditionConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py196 def validate_frozen_modality_conflicts(self) -> "ValidationSample":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py310def _get_strategy_discriminator(v: dict | TrainingStrategyConfigBase) -> str:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py425 def validate_preprocessed_data_root(cls, v: str) -> str:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py586 def migrate_deprecated_guidance_fields(cls, data: object) -> object:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py640 def validate_reference_videos(cls, v: list[str] | None, info: ValidationInfo) -> list[str] | None:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py677 def validate_output_modality_requirements(self) -> "ValidationConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py819 def _validate_data_dirs_exist(self) -> None:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/config.py830 def validate_strategy_compatibility(self) -> "LtxTrainerConfig":CODE
LOWpackages/ltx-trainer/src/ltx_trainer/gemma_8bit.py137def _suppress_accelerate_memory_warnings() -> Generator[None, None, None]:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py308 def _get_resolution_bucket_for_item(self, media_tensor: torch.Tensor) -> tuple[int, int, int]:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py385def _compute_temporal_subsample_indices(num_frames: int, factor: int) -> list[int]:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py1318def compute_scaled_resolution_buckets(CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py1377def _build_sharded_dataloader(CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py351def load_embeddings_processor(CODE
LOW⚡…kages/ltx-trainer/src/ltx_trainer/validation_runner.py372 def _encode_conditioning_media(self) -> list[CachedSampleMedia]:CODE
LOW⚡…kages/ltx-trainer/src/ltx_trainer/validation_runner.py378 def _cond_needs_video_encoder(cond: object) -> bool:CODE
LOW⚡…kages/ltx-trainer/src/ltx_trainer/validation_runner.py383 def _cond_needs_audio_encoder(cond: object) -> bool:CODE
LOW⚡…kages/ltx-trainer/src/ltx_trainer/validation_runner.py688 def _apply_video_conditionings(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py184 def _validate_sample_geometry(self) -> None:CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py418 def _encode_sample_conditions(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py520 def _encode_temporal_condition_media(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py733 def _apply_audio_conditionings(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py825 def _apply_reference_side_by_side(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1089 def _load_and_downsample_mask(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1217 def _resolve_video_guider_params(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1239 def _resolve_audio_guider_params(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1261 def _modality_from_latent_state(state: LatentState, context: Tensor, sigma: Tensor) -> Modality:CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1277 def _build_stg_perturbation_config(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1296 def _build_modality_perturbation_config(CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py1422def _concatenate_videos_side_by_side(left_video: Tensor, right_video: Tensor) -> Tensor:CODE
LOW…kages/ltx-trainer/src/ltx_trainer/timestep_samplers.py122 def _get_shift_for_sequence_length(CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py451 def _collect_trainable_params(self) -> None:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py619 def _prepare_models_for_training(self) -> None:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py825 def _offloaded_optimizer_state(self) -> Iterator[None]:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py1080 def _build_checkpoint_metadata(self) -> dict[str, str]:CODE
LOW…r/src/ltx_trainer/training_strategies/base_strategy.py201 def _create_per_token_timesteps(conditioning_mask: Tensor, sampled_sigma: Tensor) -> Tensor:CODE
LOW…r/src/ltx_trainer/training_strategies/base_strategy.py217 def _create_first_frame_conditioning_mask(CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py166 def validate_at_least_one_generated(self) -> "FlexibleStrategyConfig":CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py175 def validate_audio_intrinsic_regions(self) -> "FlexibleStrategyConfig":CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py320 def _infer_reference_scale_factors_from_config(self) -> tuple[int | None, int | None]:CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py464 def _apply_intrinsic_condition(CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py544 def _compute_spatial_crop_mask(CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py597 def _apply_reference_condition(CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py704 def _infer_temporal_scale_factor(cond_num_frames: int, target_num_frames: int) -> int:CODE
LOW…/src/ltx_trainer/training_strategies/video_to_video.py261 def _infer_reference_downscale_factor(CODE
LOW⚡…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py32def read_lora_reference_downscale_factor(lora_path: str) -> int:CODE
LOW⚡…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py43def read_lora_reference_temporal_scale_factor(lora_path: str) -> int:CODE
LOW⚡…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py54def downsample_mask_video_to_latent(CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py95def _reference_prefix_latent_frames(CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py109def append_ic_lora_reference_video_conditionings( # noqa: PLR0912, PLR0913CODE
260 more matches not shown…
Over-Commented Block195 hits · 107 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/configs/v2a_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml21# preprocessed_data_root/COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml41 # VAE components. Leave these out for a unified checkpoint — they default to model_path.COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml61 rank: 32COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml81 # - audio_attn2.to_k, audio_attn2.to_q, audio_attn2.to_v, audio_attn2.to_out.0 (audio cross-attention to text)COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml121 # modality-specific configuration blocks.COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml141 latents_dir: "audio_latents"COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml181# Acceleration ConfigurationCOMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml201COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml301 generate_audio: trueCOMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml321 keep_last_n: -1COMMENT
LOWpackages/ltx-trainer/configs/v2a_lora.yaml341# -----------------------------------------------------------------------------COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml21# Dataset structure:COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml41 # VAE components. Leave these out for a unified checkpoint — they default to model_path.COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml61 rank: 32COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml81 # - audio_attn2.to_k, audio_attn2.to_q, audio_attn2.to_v, audio_attn2.to_out.0 (audio cross-attention to text)COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml121 latents_dir: "audio_latents"COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml161 scheduler_type: "linear"COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml181 quantization: nullCOMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml301 # Skip validation at the beginning of training (step 0)COMMENT
LOWpackages/ltx-trainer/configs/audio_extend_lora.yaml321# -----------------------------------------------------------------------------COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml21# Dataset structure:COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml41 # VAE components. Leave these out for a unified checkpoint — they default to model_path.COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml61 rank: 32COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml81 # - audio_attn2.to_k, audio_attn2.to_q, audio_attn2.to_v, audio_attn2.to_out.0 (audio cross-attention to text)COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml121 latents_dir: "audio_latents"COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml161 scheduler_type: "linear"COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml181 quantization: nullCOMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml301 # Skip validation at the beginning of training (step 0)COMMENT
LOWpackages/ltx-trainer/configs/audio_suffix_lora.yaml321# -----------------------------------------------------------------------------COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml21# ├── latents/ # Video latents (VAE-encoded videos)COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml41 # A split pack's transformer carries no VAE weights, so there they are required:COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml61COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml81 # - audio_ff.net.0.proj, audio_ff.net.2 (audio feed-forward)COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml101 # - "audio_ff.net.2"COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml121 latents_dir: "latents"COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml161 # Options: "constant", "linear", "cosine", "cosine_with_restarts", "polynomial"COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml181 # Options: null (none), "int8-quanto", "int4-quanto", "int2-quanto", "fp8-quanto", "fp8uz-quanto"COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml201 preprocessed_data_root: "/path/to/preprocessed/data"COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml301COMMENT
LOWpackages/ltx-trainer/configs/a2v_lora.yaml321COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml21# -----------------------------------------------------------------------------COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml41 training_mode: "lora"COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml61 # Can help with regularization if overfitting occursCOMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml81 # (Q from audio, K/V from video - allows audio to attend to video features)COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml101# This configuration trains both video and audio generation from text prompts.COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml121 # Directory name (within preprocessed_data_root) containing audio latentsCOMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml161# -----------------------------------------------------------------------------COMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml181 offload_optimizer_during_validation: falseCOMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml281 # Independent of training_strategy.audio.is_generated - you can generate audioCOMMENT
LOWpackages/ltx-trainer/configs/t2v_lora.yaml301 # Precision to use when saving checkpoint weightsCOMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml1# =============================================================================COMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml21# Dataset metadata columns: video, reference_video, captionCOMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml41 # audio_vae_path: "path/to/vae/ltx-2.5-audio-vae-bf16.safetensors"COMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml61 # The effective scaling is alpha/rank, so alpha=rank means scaling of 1.0COMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml81 # AUDIO-VIDEO CROSS-ATTENTION MODULES (for cross-modal interaction, not used for video-only IC-LoRA):COMMENT
135 more matches not shown…
Modern Structural Boilerplate87 hits · 88 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py470 def _setup_lora(self) -> None:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py777 def _setup_accelerator(self) -> None:CODE
LOW…rainer/src/ltx_trainer/training_strategies/__init__.py25__all__ = [CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/distilled_mgpu.py41logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/alpha_gen.py58logger = logging.getLogger(__name__)CODE
LOW…-pipelines/src/ltx_pipelines/ti2vid_two_stages_mgpu.py43logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/__init__.py52__all__ = [CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dubit.py84__all__ = ["DubItPipeline", "main", "patchify_dubit_audio_reference_latent"]CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dfr_stages.py59logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/hdr_ic_lora.py54logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dfr_mgpu.py46logger = logging.getLogger(__name__)CODE
LOW…pelines/src/ltx_pipelines/ti2vid_two_stages_hq_mgpu.py45logger = logging.getLogger(__name__)CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/a2vid_two_stage.py80logger = logging.getLogger(__name__)CODE
LOW⚡…ltx-pipelines/src/ltx_pipelines/multigpu/controller.py48logger = logging.getLogger(__name__)CODE
LOW⚡…ages/ltx-pipelines/src/ltx_pipelines/multigpu/fleet.py33logger = logging.getLogger(__name__)CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/multigpu/__init__.py22__all__ = [CODE
LOW…pipelines/src/ltx_pipelines/multigpu/gemma_builders.py37logger = logging.getLogger(__name__)CODE
LOW…pelines/src/ltx_pipelines/multigpu/bp_gemma_builder.py39logger = logging.getLogger(__name__)CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/__init__.py40__all__ = [CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/utils/args.py30logger = logging.getLogger(__name__)CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/utils/blocks.py123logger = logging.getLogger(__name__)CODE
LOW…kages/ltx-pipelines/src/ltx_pipelines/utils/helpers.py60logger = logging.getLogger(__name__)CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/samplers.py23logger = logging.getLogger(__name__)CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py29logger = logging.getLogger(__name__)CODE
LOW…/ltx-pipelines/src/ltx_pipelines/utils/media_io/exr.py20logger = logging.getLogger(__name__)CODE
LOW…pipelines/src/ltx_pipelines/utils/media_io/__init__.py46__all__ = [CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/decode.py30logger = logging.getLogger(__name__)CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/chunks/planner.py29logger = logging.getLogger(__name__)CODE
LOW…kages/ltx-pipelines/src/ltx_pipelines/chunks/layout.py13logger = logging.getLogger(__name__)CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/chunks/__init__.py33__all__ = [CODE
LOW…kages/ltx-pipelines/src/ltx_pipelines/chunks/export.py21logger = logging.getLogger(__name__)CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/dfr_helpers/ops.py20logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/modality_tiling.py32logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/hdr.py14__all__ = [CODE
LOWpackages/ltx-core/src/ltx_core/devices.py18logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/multigpu/__init__.py11__all__ = ["DimensionTilingConfig", "ShardedSD", "TileCountConfig", "transformer", "vae"]CODE
LOW…kages/ltx-core/src/ltx_core/multigpu/gemma/__init__.py6__all__ = ["AccelerateGemmaWrapper", "load_gemma_with_device_map"]CODE
LOWpackages/ltx-core/src/ltx_core/multigpu/gemma/loader.py29logger = logging.getLogger(__name__)CODE
LOW…ltx-core/src/ltx_core/multigpu/transformer/__init__.py11__all__ = [CODE
LOW…-core/src/ltx_core/multigpu/vae/distributed_decoder.py50logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/multigpu/vae/__init__.py5__all__ = ["DistributedVideoDecoder"]CODE
LOW…/ltx-core/src/ltx_core/text_encoders/gemma/__init__.py30__all__ = [CODE
LOW…-core/src/ltx_core/text_encoders/gemma/gemma_assets.py32logger = logging.getLogger(__name__)CODE
LOW…e/text_encoders/gemma/encoders/encoder_configurator.py25logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/guidance/__init__.py10__all__ = [CODE
LOWpackages/ltx-core/src/ltx_core/color/__init__.py36__all__ = [CODE
LOWpackages/ltx-core/src/ltx_core/color/hlg.py33logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/conditioning/__init__.py16__all__ = [CODE
LOW…s/ltx-core/src/ltx_core/conditioning/types/__init__.py11__all__ = [CODE
LOWpackages/ltx-core/src/ltx_core/quantization/__init__.py10__all__ = [CODE
LOW…x-core/src/ltx_core/quantization/blockwise/__init__.py11__all__ = ["build_fp6_policy", "build_fp8_policy"]CODE
LOW…s/ltx-core/src/ltx_core/quantization/nvfp4/__init__.py13__all__ = [CODE
LOWpackages/ltx-core/src/ltx_core/loader/__init__.py38__all__ = [CODE
LOW…ltx-core/src/ltx_core/block_streaming/block_fetcher.py14logger = logging.getLogger(__name__)CODE
LOW…ages/ltx-core/src/ltx_core/block_streaming/__init__.py16__all__ = [CODE
LOW…kages/ltx-core/src/ltx_core/block_streaming/builder.py49logger = logging.getLogger(__name__)CODE
LOWpackages/ltx-core/src/ltx_core/model/__init__.py6__all__ = [CODE
LOW…es/ltx-core/src/ltx_core/model/transformer/__init__.py13__all__ = [CODE
LOW…kages/ltx-core/src/ltx_core/model/transformer/model.py22logger = logging.getLogger(__name__)CODE
LOW…kages/ltx-core/src/ltx_core/model/transformer/model.py416 def set_gradient_checkpointing(self, enable: bool) -> None:CODE
27 more matches not shown…
AI Structural Patterns65 hits · 64 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/scripts/split_scenes.py153CODE
LOWpackages/ltx-trainer/scripts/split_scenes.py310CODE
LOWpackages/ltx-trainer/scripts/caption_videos.py352CODE
LOWpackages/ltx-trainer/scripts/process_dataset.py55CODE
LOWpackages/ltx-trainer/scripts/process_dataset.py315CODE
LOWpackages/ltx-trainer/scripts/decode_latents.py313CODE
LOWpackages/ltx-trainer/scripts/process_captions.py233CODE
LOWpackages/ltx-trainer/scripts/process_captions.py399CODE
LOWpackages/ltx-trainer/scripts/serve_captioner.py99CODE
LOWpackages/ltx-trainer/src/ltx_trainer/datasets.py18CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py396CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py837CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py1400CODE
LOWpackages/ltx-trainer/src/ltx_trainer/model_loader.py408CODE
LOW…-pipelines/src/ltx_pipelines/keyframe_interpolation.py70CODE
LOW…-pipelines/src/ltx_pipelines/keyframe_interpolation.py152CODE
LOW…es/ltx-pipelines/src/ltx_pipelines/ti2vid_one_stage.py77CODE
LOW…es/ltx-pipelines/src/ltx_pipelines/ti2vid_one_stage.py143CODE
LOW…tx-pipelines/src/ltx_pipelines/ti2vid_two_stages_hq.py82CODE
LOW…tx-pipelines/src/ltx_pipelines/ti2vid_two_stages_hq.py185CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/distilled_mgpu.py143CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/ti2vid_two_stages.py115CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/ti2vid_two_stages.py312CODE
LOW…s/ltx-pipelines/src/ltx_pipelines/ti2vid_two_stages.py366CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/retake.py78CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/retake.py151CODE
LOW…-pipelines/src/ltx_pipelines/ti2vid_two_stages_mgpu.py163CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/ic_lora.py296CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/ic_lora.py583CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/ic_lora.py669CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dubit.py114CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dubit.py318CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dubit.py359CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/dfr_pipeline.py113CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/dfr_pipeline.py244CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/distilled.py166CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/distilled.py357CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/distilled.py403CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/dfr_mgpu.py133CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/a2vid_two_stage.py117CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/a2vid_two_stage.py323CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/a2vid_two_stage.py378CODE
LOW…ckages/ltx-pipelines/src/ltx_pipelines/utils/blocks.py304CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/samplers.py207CODE
LOW…tx-core/src/ltx_core/multigpu/transformer/attention.py90CODE
LOW…c/ltx_core/text_encoders/gemma/embeddings_connector.py95CODE
LOWpackages/ltx-core/src/ltx_core/color/hlg.py258CODE
LOW…kages/ltx-core/src/ltx_core/block_streaming/builder.py84CODE
LOW…s/ltx-core/src/ltx_core/model/transformer/attention.py96CODE
LOW…s/ltx-core/src/ltx_core/model/transformer/attention.py141CODE
LOW…s/ltx-core/src/ltx_core/model/transformer/attention.py166CODE
LOW…s/ltx-core/src/ltx_core/model/transformer/attention.py191CODE
LOW…kages/ltx-core/src/ltx_core/model/transformer/model.py43CODE
LOW…ckages/ltx-core/src/ltx_core/model/transformer/rope.py40CODE
LOWpackages/ltx-core/src/ltx_core/model/upsampler/model.py25CODE
LOW…kages/ltx-core/src/ltx_core/model/audio_vae/vocoder.py317CODE
LOW…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py67CODE
LOW…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py284CODE
LOW…s/ltx-core/src/ltx_core/model/video_vae/convolution.py11CODE
LOW…ore/src/ltx_core/model/video_vae/conv_video_decoder.py179CODE
5 more matches not shown…
Hallucination Indicators6 hits · 60 pts
SeverityFileLineSnippetContext
CRITICAL…ges/ltx-pipelines/src/ltx_pipelines/utils/denoisers.py22from ltx_pipelines.utils.helpers import modality_from_latent_stateCODE
CRITICAL…ges/ltx-pipelines/src/ltx_pipelines/utils/gpu_model.py8from ltx_pipelines.utils.helpers import cleanup_memoryCODE
CRITICAL…ages/ltx-pipelines/src/ltx_pipelines/utils/samplers.py19from ltx_pipelines.utils.helpers import post_process_latent, timesteps_from_maskCODE
CRITICAL…tx-pipelines/src/ltx_pipelines/chunks/conditionings.py29from ltx_pipelines.utils.helpers import combined_image_conditioningsCODE
CRITICAL…kages/ltx-pipelines/src/ltx_pipelines/chunks/layout.py11from ltx_pipelines.utils.helpers import snap_frames_to_gridCODE
CRITICAL…es/ltx-pipelines/src/ltx_pipelines/chunks/operators.py32from ltx_pipelines.utils.helpers import decode_keyframes_from_slotsCODE
Deep Nesting61 hits · 52 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/scripts/compute_reference.py142CODE
LOWpackages/ltx-trainer/scripts/caption_videos.py77CODE
LOWpackages/ltx-trainer/scripts/caption_videos.py284CODE
LOWpackages/ltx-trainer/scripts/process_captions.py233CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py15CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py113CODE
LOWpackages/ltx-trainer/src/ltx_trainer/quantization.py171CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py396CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py840CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py1084CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py418CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py688CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py733CODE
LOW…kages/ltx-trainer/src/ltx_trainer/validation_runner.py825CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py134CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py724CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py825CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py187CODE
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py464CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/ic_lora.py669CODE
LOWpackages/ltx-pipelines/src/ltx_pipelines/utils/args.py215CODE
LOW…ages/ltx-pipelines/src/ltx_pipelines/utils/samplers.py207CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py32CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py329CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py76CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py345CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/decode.py137CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/decode.py304CODE
LOWpackages/ltx-core/src/ltx_core/multigpu/sharded_sd.py123CODE
LOWpackages/ltx-core/src/ltx_core/multigpu/gemma/loader.py70CODE
LOW…-core/src/ltx_core/text_encoders/gemma/gemma_assets.py253CODE
LOWpackages/ltx-core/src/ltx_core/loader/sft_loader.py20CODE
LOWpackages/ltx-core/src/ltx_core/block_streaming/disk.py88CODE
LOW…kages/ltx-core/src/ltx_core/block_streaming/builder.py471CODE
LOW…ore/src/ltx_core/model/video_vae/model_configurator.py519CODE
LOW…ore/src/ltx_core/model/video_vae/conv_video_decoder.py64CODE
LOW…ore/src/ltx_core/model/video_vae/conv_video_decoder.py386CODE
LOW…ges/ltx-core/src/ltx_core/model/video_vae/video_vae.py41CODE
LOW…rc/ltx_core/model/video_vae/diffusion_video_decoder.py1004CODE
LOW…rc/ltx_core/model/video_vae/diffusion_video_decoder.py1345CODE
LOW…rc/ltx_core/model/video_vae/memory_efficient_decode.py541CODE
LOW…-core/src/ltx_core/model/video_vae/diffusion_tiling.py261CODE
LOW…e/model/video_vae/transformer/fallback_na/triton_na.py21CODE
LOW…model/video_vae/transformer/fallback_na/joint_eager.py428CODE
LOWpackages/ltx-kernels/setup.py153CODE
LOWpackages/ltx-kernels/setup.py194CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py91CODE
LOW…rnels/src/ltx_kernels/compile/passes/grouped_upcast.py43CODE
LOWpackages/ltx-kernels/src/ltx_kernels/vae/na_attn_dsl.py88CODE
LOWpackages/ltx-kernels/src/ltx_kernels/vae/na_attn_dsl.py113CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py399CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1084CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1318CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1524CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1602CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1661CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1688CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py1706CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/block_fna_dsl.py2715CODE
LOW…kages/ltx-kernels/src/ltx_kernels/vae/fna_attn_core.py314CODE
1 more matches not shown…
Excessive Try-Catch Wrapping35 hits · 36 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/scripts/compute_reference.py103 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/compute_reference.py228 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/split_scenes.py283 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/caption_videos.py179 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/caption_videos.py346 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/train.py54 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/decode_latents.py137 except Exception as e:CODE
LOWpackages/ltx-trainer/scripts/decode_latents.py186 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py41 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py80 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py109 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py173 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/datasets.py274 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py247 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/process_videos.py1261 except Exception:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/captioning.py274 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/captioning.py376 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py586 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py603 except Exception as e:CODE
LOWpackages/ltx-trainer/src/ltx_trainer/trainer.py1047 except Exception:CODE
LOW⚡…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py38 except Exception as e:CODE
LOW⚡…ckages/ltx-pipelines/src/ltx_pipelines/iclora_utils.py49 except Exception as e:CODE
LOW…ltx-pipelines/src/ltx_pipelines/multigpu/controller.py262 except Exception:CODE
LOW…ltx-pipelines/src/ltx_pipelines/multigpu/controller.py355 except Exception:CODE
LOW…ges/ltx-pipelines/src/ltx_pipelines/utils/constants.py152 except Exception:CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py358 except Exception as e:CODE
LOW…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py364 except Exception as e:CODE
LOWpackages/ltx-core/src/ltx_core/devices.py159 except Exception:CODE
LOWpackages/ltx-core/src/ltx_core/color/hlg.py218 except Exception as e:CODE
LOWpackages/ltx-core/src/ltx_core/color/hlg.py223 except Exception as e:CODE
LOW…ltx-core/src/ltx_core/block_streaming/block_fetcher.py80 except Exception as exc:CODE
LOW…s/ltx-core/src/ltx_core/model/transformer/attention.py36except Exception:CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py81 except Exception as error:CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py132 except Exception as error:CODE
LOWpackages/ltx-kernels/src/ltx_kernels/vae/aot.py136 except Exception as exc:CODE
Structural Annotation Overuse16 hits · 28 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/docs/custom-training-strategies.md58### Step 1: Plan Your StrategyCOMMENT
LOWpackages/ltx-trainer/docs/custom-training-strategies.md74### Step 2: Extend Data Preprocessing (If Needed)COMMENT
LOWpackages/ltx-trainer/docs/custom-training-strategies.md130### Step 3: Create the Strategy ConfigurationCOMMENT
LOWpackages/ltx-trainer/docs/custom-training-strategies.md198### Step 4: Implement the Strategy ClassCOMMENT
LOWpackages/ltx-trainer/docs/custom-training-strategies.md324### Step 5: Register the StrategyCOMMENT
LOWpackages/ltx-trainer/docs/custom-training-strategies.md369### Step 6: Create a Configuration FileCOMMENT
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py361 # Step 1: Load and patchify latentsCOMMENT
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py369 # Step 2: Get text embeddingsCOMMENT
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py374 # Step 3: Initialize noise, timesteps, and loss mask based on is_generated flagCOMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py387 # Step 4: Generate positionsCOMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py404 # Step 5: Apply conditions (intrinsic first, then extrinsic)COMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py431 # Step 6: Build ModalityCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py448 # Step 1: Crop first to avoid exceeding target dimensionsCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py453 # Step 2: Calculate padding needed for time and frequency dimensionsCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py457 # Step 3: Apply padding if neededCOMMENT
LOW…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py469 # Step 4: Final safety crop to ensure exact target shapeCOMMENT
Docstring Block Structure5 hits · 25 pts
SeverityFileLineSnippetContext
HIGHpackages/ltx-trainer/src/ltx_trainer/gemma_8bit.py61Quantize a built text encoder's Gemma backbone to 8-bit and move it to ``device``. Args: text_encoder: EncodSTRING
HIGH…kages/ltx-trainer/src/ltx_trainer/timestep_samplers.py105Sample timesteps for a specific batch tensor. Args: batch: Input tensor of shape (batch_size, seq_leSTRING
HIGH…rainer/src/ltx_trainer/training_strategies/__init__.py42Factory function to create the appropriate training strategy. The strategy is determined by the `name` field in the STRING
HIGH…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py43Write EXR frames plus a BT.2020/HLG master. One pass over ``video`` feeds two independent sinks: * **HLG** — alwSTRING
HIGH…x-pipelines/src/ltx_pipelines/utils/media_io/encode.py241Encode RGB frames to video; SDR H.264 or HDR EXR+HLG when ``color_space`` is set. Args: video: RGB frames asSTRING
Empty Error Handlers5 hits · 25 pts
SeverityFileLineSnippetContext
HIGHpackages/ltx-trainer/src/ltx_trainer/captioning.py414except (json.JSONDecodeError, ValueError):CODE
HIGHpackages/ltx-trainer/src/ltx_trainer/captioning.py423except (json.JSONDecodeError, ValueError):CODE
HIGHpackages/ltx-trainer/src/ltx_trainer/captioning.py261except subprocess.CalledProcessError:CODE
HIGH…ltx-pipelines/src/ltx_pipelines/multigpu/controller.py137except (SymmetricRunnerError, AsymmetricRunnerError):CODE
HIGH…/src/ltx_core/model/video_vae/transformer/compiling.py59except ImportError:CODE
Verbosity Indicators10 hits · 20 pts
SeverityFileLineSnippetContext
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py361 # Step 1: Load and patchify latentsCOMMENT
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py369 # Step 2: Get text embeddingsCOMMENT
LOW⚡…rainer/src/ltx_trainer/training_strategies/flexible.py374 # Step 3: Initialize noise, timesteps, and loss mask based on is_generated flagCOMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py387 # Step 4: Generate positionsCOMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py404 # Step 5: Apply conditions (intrinsic first, then extrinsic)COMMENT
LOW…rainer/src/ltx_trainer/training_strategies/flexible.py431 # Step 6: Build ModalityCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py448 # Step 1: Crop first to avoid exceeding target dimensionsCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py453 # Step 2: Calculate padding needed for time and frequency dimensionsCOMMENT
LOW⚡…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py457 # Step 3: Apply padding if neededCOMMENT
LOW…ges/ltx-core/src/ltx_core/model/audio_vae/audio_vae.py469 # Step 4: Final safety crop to ensure exact target shapeCOMMENT
Cross-File Repetition3 hits · 15 pts
SeverityFileLineSnippetContext
HIGH…s/ltx-pipelines/src/ltx_pipelines/ti2vid_two_stages.py0validate inputs, encode prompts/reference, and resolve frame count / tiling before chunk planning.STRING
HIGHpackages/ltx-pipelines/src/ltx_pipelines/distilled.py0validate inputs, encode prompts/reference, and resolve frame count / tiling before chunk planning.STRING
HIGHpackages/ltx-pipelines/src/ltx_pipelines/dubit.py0validate inputs, encode prompts/reference, and resolve frame count / tiling before chunk planning.STRING
AI Slop Vocabulary3 hits · 14 pts
SeverityFileLineSnippetContext
MEDIUM⚡packages/ltx-trainer/configs/audio_extend_lora.yaml19# - Create seamless audio extension modelsCOMMENT
MEDIUM⚡packages/ltx-trainer/configs/audio_suffix_lora.yaml19# - Create seamless audio extension modelsCOMMENT
MEDIUM⚡packages/ltx-trainer/configs/video_extend_lora.yaml17# - Create seamless video extension models with optional audioCOMMENT
Self-Referential Comments3 hits · 9 pts
SeverityFileLineSnippetContext
MEDIUMpackages/ltx-trainer/src/ltx_trainer/hf_hub_utils.py45 # Create a single temporary directory for all filesCOMMENT
MEDIUM…ckages/ltx-core/src/ltx_core/components/patchifiers.py108 # Create a tensor containing the size of a single patch:COMMENT
MEDIUM…s/ltx-core/src/ltx_core/model/video_vae/convolution.py124 # Define the size of the channels after the first convolutionCOMMENT
Modern AI Meta-Vocabulary4 hits · 8 pts
SeverityFileLineSnippetContext
MEDIUMpackages/ltx-trainer/AGENTS.md58│ ├── trainer.py # Main training orchestration with AccelerateCODE
MEDIUMpackages/ltx-trainer/CLAUDE.md58│ ├── trainer.py # Main training orchestration with AccelerateCODE
MEDIUMpackages/ltx-trainer/scripts/caption_videos.py25 # Enable Qwen3 chain-of-thought (slower, more detail)STRING
MEDIUMpackages/ltx-trainer/src/ltx_trainer/captioning.py400 # Thinking models (e.g. Qwen3-Omni-*-Thinking) emit the reasoning traceCOMMENT
Redundant / Tautological Comments7 hits · 8 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml234 # Set these to match --reference-downscale-factor / --reference-temporal-scale-factorCOMMENT
LOWpackages/ltx-trainer/configs/v2v_ic_lora.yaml249 # Set these to match --reference-downscale-factor / --reference-temporal-scale-factorCOMMENT
LOWpackages/ltx-trainer/scripts/decode_latents.py152 # Check if audio VAE is loadedCOMMENT
LOWpackages/ltx-trainer/scripts/decode_latents.py205 # Check if latents need reshaping (old patchified format)COMMENT
LOWpackages/ltx-trainer/src/ltx_trainer/quantization.py89 # Check if model has transformer_blocks for block-by-block quantizationCOMMENT
LOWpackages/ltx-trainer/src/ltx_trainer/datasets.py289 # Check if latents are in legacy patchified format [seq_len, C]COMMENT
LOW…ore/src/ltx_core/model/video_vae/conv_video_decoder.py447 # Check if current temporal slice overlaps with previous temporal sliceSTRING
Cross-Language Confusion1 hit · 5 pts
SeverityFileLineSnippetContext
HIGH…rc/ltx_core/model/video_vae/memory_efficient_decode.py337 interior.mul_(1 + scale2).add_(shift2) # type: ignore[possibly-undefined]CODE
Fake / Example Data5 hits · 5 pts
SeverityFileLineSnippetContext
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py106 pattern_placeholders = {node.target: node for node in pattern_graph.nodes if node.op == "placeholder"}CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py136 replacement_placeholders = [node for node in replacement_graph.nodes if node.op == "placeholder"]CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py148 if pattern_node.op not in ("placeholder", "output") and isinstance(graph_node, Node)CODE
LOW…rnels/src/ltx_kernels/compile/subgraph_rewrite_pass.py223 for placeholder in (node for node in self._pattern_gm.graph.nodes if node.op == "placeholder"):CODE
LOW…rnels/src/ltx_kernels/compile/passes/grouped_upcast.py17 if dtype is not torch.bfloat16 or not isinstance(source, Node) or source.op != "placeholder":CODE
Overly Generic Function Names1 hit · 0 pts
SeverityFileLineSnippetContext
LOWpackages/ltx-trainer/src/ltx_trainer/gpu_utils.py33 def my_function():STRING