Lightweight Image Video Action Generation Inference Framework
This report summarizes scanner evidence for ModelTC/LightX2V, a Python project with 2,897 GitHub stars. SynthScan examined 373,548 lines of code across 2998 source files, recording 4201 pattern matches distributed across 24 syntactic categories. The legacy adjusted pattern signal is 16.7 (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? 2110 finding(s).
Are imports declared and package/API names verified? 891 finding(s).
Are wrappers, guards, and repeated structures necessary? 2664 finding(s).
Which changes need focused human review? 4201 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 4201 distinct pattern matches across 24 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 | tools/preprocess/dwpose_onnx.py | 28 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 30 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 143 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 145 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 253 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 255 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 411 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | tools/preprocess/dwpose_onnx.py | 413 | # ----------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v_platform/ops/attn/intel_xpu/xpu_flash_attn.py | 110 | # ── normalise 4-D input [B, S, H, D] → [B*S, H, D] ────────────────── | COMMENT |
| MEDIUM | lightx2v_platform/ops/attn/intel_xpu/xpu_flash_attn.py | 123 | # ── fast single-sequence path ───────────────────────────────────── | COMMENT |
| MEDIUM | lightx2v_platform/ops/attn/intel_xpu/xpu_flash_attn.py | 130 | # ── varlen path: one SDPA call per sequence in the batch ────────────── | COMMENT |
| MEDIUM | lightx2v_platform/ops/mm/template.py | 122 | # ========================= | COMMENT |
| MEDIUM | lightx2v_platform/ops/mm/template.py | 124 | # ========================= | COMMENT |
| MEDIUM | lightx2v_platform/ops/mm/template.py | 410 | # ========================= | COMMENT |
| MEDIUM | lightx2v_platform/ops/mm/template.py | 412 | # ========================= | COMMENT |
| MEDIUM | …x2v/models/video_encoders/hf/ltx2/audio_vae/vocoder.py | 39 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | …x2v/models/video_encoders/hf/ltx2/audio_vae/vocoder.py | 42 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 717 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 721 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 991 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 994 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 1004 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | lightx2v/models/networks/worldmirror/model.py | 1006 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 36 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 40 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 155 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 158 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 188 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 190 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 286 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 289 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 337 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 339 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 374 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 376 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 467 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 478 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 566 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 568 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 966 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | lightx2v/models/networks/worldmirror/model.py | 968 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 425 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 431 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 465 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 472 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 499 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM | …ls/networks/worldmirror/weights/transformer_weights.py | 501 | # ------------------------------------------------------------------ | COMMENT |
| MEDIUM⚡ | …v/models/networks/worldmirror/utils/inference_utils.py | 183 | # ============================================================ | COMMENT |
| MEDIUM⚡ | …v/models/networks/worldmirror/utils/inference_utils.py | 185 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 37 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 39 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 118 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 120 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 341 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 343 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 528 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 530 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 810 | # ============================================================ | COMMENT |
| MEDIUM | …v/models/networks/worldmirror/utils/inference_utils.py | 812 | # ============================================================ | COMMENT |
| MEDIUM | …els/networks/flux2/infer/pipefusion/pipeline_driver.py | 39 | # ================================================================== | COMMENT |
| 326 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/cache_minimax_h3_adaln/builder.py | 169 | def build_persistent_adaln_cache(config) -> Path: | CODE |
| LOW | tools/preprocess/retarget_pose.py | 357 | def fix_lack_keypoints_use_sym(skeleton): | CODE |
| LOW | tools/preprocess/process_pipepline.py | 72 | def _get_valid_body_keypoint_mask(cls, pose_meta, keypoint_indices=None): | CODE |
| LOW | tools/preprocess/process_pipepline.py | 90 | def _count_valid_body_keypoints(cls, pose_meta): | CODE |
| LOW | tools/preprocess/process_pipepline.py | 105 | def _trim_tail_invalid_frames(frames, pose_metas, pose_valid_flags, mode_name): | CODE |
| LOW | tools/preprocess/utils.py | 27 | def skip_replace_frame_outputs(frame): | CODE |
| LOW | tools/preprocess/pose2d_utils.py | 740 | def keypoints_from_regression(regression_preds, center, scale, img_size): | CODE |
| LOW | tools/preprocess/pose2d_utils.py | 932 | def multilabel_classification_accuracy(pred, gt, mask, thr=0.5): | CODE |
| LOW | tools/profile/analyze_torch_trace_gap.py | 127 | def _collect_gpu_activity_intervals(events: Iterable[dict]) -> list[Interval]: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 177 | def derive_replicated_prefix_split(total_sequence: int, conditioner_tokens: int, sp_size: int) -> tuple[int, int]: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 289 | def _packed_tensor_description(name: str, shape: list[int], quant_scheme: str | None) -> dict[str, Any]: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 562 | def enumerate_sparse_candidates( | CODE |
| LOW | tools/benchmarks/sp_bench.py | 866 | def _sparse_attention_operator(backend: str, keep_ratio: float) -> Any: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1008 | def _result_correctness_policy(case: dict[str, Any], candidate: dict[str, Any], *, allow_missing: bool = False) -> dict[ | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1025 | def _result_correctness_tolerance(case: dict[str, Any], candidate: dict[str, Any]) -> float: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1167 | def _bulk_reference_candidate(candidate: dict[str, Any]) -> dict[str, Any]: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1173 | def _reference_validation_candidates(case: dict[str, Any], candidates: list[dict[str, Any]]) -> list[dict[str, Any]]: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1853 | def _validate_report_record_identities( | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1912 | def validate_existing_records( | CODE |
| LOW | tools/benchmarks/sp_bench.py | 2515 | def render_diagnostic_markdown(report: dict[str, Any]) -> str: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 2600 | def diagnostic_candidates_from_report(path: Path, suite: dict[str, Any]) -> dict[str, str]: | CODE |
| LOW | tools/benchmarks/operator_backends.py | 654 | def _prepare_dynamic_sparse_replay( | CODE |
| LOW | tools/benchmarks/operator_backends.py | 693 | def _prepare_production_sparse_replay( | CODE |
| LOW | tools/benchmarks/operator_backends.py | 755 | def _prepare_production_dense_replay( | CODE |
| LOW | tools/benchmarks/operator_backends.py | 815 | def _prepare_production_dense( | CODE |
| LOW | tools/benchmarks/operator_backends.py | 1011 | def build_backend_catalog_report( | CODE |
| LOW | tools/benchmarks/operator_bench.py | 273 | def parse_backend_assignments(values: list[str]) -> dict[str, list[str]]: | CODE |
| LOW | tools/benchmarks/operator_bench.py | 648 | def build_recommendation_report( | CODE |
| LOW | tools/extract/lora_merger.py | 350 | def _find_source_key_from_diff(diff_key: str, source_weights: Dict[str, torch.Tensor]) -> Optional[str]: | CODE |
| LOW | tools/postprocess/postprocess_paint.py | 15 | def get_postprocess_paint_parser(): | CODE |
| LOW | tools/postprocess/torchvision_fix.py | 7 | def fix_torchvision_functional_tensor(): | CODE |
| LOW | tools/postprocess/paint_pipeline.py | 90 | def _load_texture_gen_pipeline_module(hy3dpaint_root: str) -> ModuleType: | CODE |
| LOW | tools/postprocess/paint_pipeline.py | 110 | def resolve_paint_output_paths(save_path: str) -> tuple[str, str]: | CODE |
| LOW | tools/postprocess/paint_pipeline.py | 120 | def _resolve_local_paint_hf_root(model_path: str) -> str: | CODE |
| LOW | tools/postprocess/paint_pipeline.py | 131 | def _paint_model_source_overlay(model_path: str, hy3dpaint_root: str) -> Iterator[str]: | CODE |
| LOW | tools/convert/export_dummy_meta.py | 43 | def write_dummy_meta_safetensors(tensor_meta: dict, output_path: str, source_filename: str = ""): | CODE |
| LOW | tools/convert/seko_talk_converter.py | 132 | def checkpoint_to_safetensors(ckpt_path: Path, output_path: Path, description: str = ""): | CODE |
| LOW | lightx2v_platform/ops/rope/rope_template.py | 70 | def load_state_dict_from_disk(self, block_index, adapter_block_index=None): | CODE |
| LOW | lightx2v_platform/ops/rope/biren_supa/wan_rope.py | 53 | def prepare_cos_sin_from_freqs(freqs, grid_sizes, head_size, seq_p_group=None): | CODE |
| LOW | lightx2v_platform/ops/rope/biren_supa/wan_rope.py | 95 | def _get_seq_parallel_world_size(self): | CODE |
| LOW | lightx2v_platform/ops/rope/biren_supa/qwen_rope.py | 18 | def prepare_cos_sin_from_complex(rotary_emb): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 11 | def _load_metax_rotary_embedding_op(): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 38 | def _metax_rotary_embedding_fake(query, key, positions, freqs, head_size, split_half): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 47 | def metax_rotary_embedding_single( | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 61 | def _metax_rotary_embedding_single_fake(tensor, positions, freqs, head_size, split_half): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 66 | def _register_metax_rotary_embedding_lowering(): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 122 | def _metax_rotary_embedding_lowering(query, key, positions, freqs, head_size, split_half): | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 129 | def _metax_rotary_embedding_single_lowering(tensor, positions, freqs, head_size, split_half): | CODE |
| LOW | lightx2v_platform/ops/norm/norm_template.py | 153 | def load_state_dict_from_disk(self, block_index, adapter_block_index=None): | CODE |
| LOW | lightx2v_platform/ops/norm/norm_template.py | 327 | def load_state_dict_from_disk(self, block_index, adapter_block_index=None): | CODE |
| LOW | lightx2v_platform/ops/norm/metax_cuda/metax_rms_norm.py | 36 | def _register_metax_rms_norm_lowering(): | CODE |
| LOW | lightx2v_platform/ops/attn/biren_supa/ulysses_attn.py | 104 | def _distributed_attention_sbp(self, q, k, v, shard_seq, attention_module, seq_p_group, world_size, num_heads, head_ | CODE |
| LOW | lightx2v_platform/ops/attn/intel_xpu/xpu_sla_attn.py | 38 | def apply_intel_xpu_cute_attn( | CODE |
| LOW | lightx2v_platform/ops/attn/amd_rocm/sage_attn.py | 151 | def _force_inductor_single_thread(): | CODE |
| LOW | lightx2v_platform/ops/mm/template.py | 496 | def load_state_dict_from_disk(self, block_index, adapter_block_index=None): | CODE |
| LOW | lightx2v_platform/ops/mm/biren_supa/mm_weight.py | 172 | def _offload_to_pinned_cpu_refs(self): | CODE |
| LOW | lightx2v_platform/ops/mm/biren_supa/mm_weight.py | 183 | def _move_row_bias_after_reduce(self): | CODE |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 45 | def _get_intel_tensor_parallel_class(): | CODE |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 198 | def _get_tensor_parallel_rms_class(): | CODE |
| LOW | lightx2v_platform/ops/moe/template.py | 48 | def load_state_dict_from_disk(self, block_index, adapter_block_index=None): | CODE |
| 1174 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/preprocess/dwpose_onnx.py | 13 | CODE | |
| LOW | tools/preprocess/sam_utils.py | 13 | CODE | |
| LOW | tools/preprocess/__init__.py | 2 | CODE | |
| LOW | tools/preprocess/__init__.py | 3 | CODE | |
| LOW | tools/profile/analyze_torch_trace_gap.py | 24 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 3 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 717 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 894 | CODE | |
| LOW | tools/benchmarks/operator_backends.py | 3 | CODE | |
| LOW | tools/benchmarks/operator_backends.py | 453 | CODE | |
| LOW | tools/benchmarks/operator_backends.py | 1092 | CODE | |
| LOW | tools/benchmarks/operator_backends.py | 1093 | CODE | |
| LOW | tools/benchmarks/operator_bench.py | 3 | CODE | |
| LOW | tools/postprocess/torchvision_fix.py | 9 | CODE | |
| LOW | tools/postprocess/paint_pipeline.py | 3 | CODE | |
| LOW | tools/postprocess/paint_pipeline.py | 67 | CODE | |
| LOW | tools/convert/quant_adapter.py | 15 | CODE | |
| LOW | tools/convert/converter.py | 31 | CODE | |
| LOW | tools/convert/quant/__init__.py | 1 | CODE | |
| LOW | lightx2v_platform/__init__.py | 1 | CODE | |
| LOW | lightx2v_platform/set_ai_device.py | 5 | CODE | |
| LOW | lightx2v_platform/test/test_device.py | 8 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 3 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 7 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 8 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 9 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 10 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 12 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 13 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 18 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 19 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 26 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 27 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 28 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 29 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 30 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 31 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 37 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 38 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 39 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 40 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 43 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 44 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 45 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 49 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 50 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 51 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 52 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 54 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 55 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 56 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 57 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 59 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 64 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 65 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 66 | CODE | |
| LOW | lightx2v_platform/ops/__init__.py | 67 | CODE | |
| LOW | lightx2v_platform/ops/a2a/ascend_npu/__init__.py | 1 | CODE | |
| LOW | lightx2v_platform/ops/rope/biren_supa/__init__.py | 1 | CODE | |
| LOW | lightx2v_platform/ops/rope/biren_supa/__init__.py | 2 | CODE | |
| 792 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | lightx2v/utils/ltx2_utils.py | 0 | protocol for model loader classes that instantiates models from a configuration dictionary. | STRING |
| HIGH | …htx2v/models/video_encoders/hf/ltx2/upsampler/model.py | 0 | protocol for model loader classes that instantiates models from a configuration dictionary. | STRING |
| HIGH | lightx2v/models/input_encoders/hf/ltx2/utils.py | 0 | protocol for model loader classes that instantiates models from a configuration dictionary. | STRING |
| HIGH | …video_encoders/hf/ltx2/video_vae/timestep_embedding.py | 0 | this matches the implementation in denoising diffusion probabilistic models: create sinusoidal timestep embeddings. args | STRING |
| HIGH | lightx2v/models/networks/ltx2/infer/utils.py | 0 | this matches the implementation in denoising diffusion probabilistic models: create sinusoidal timestep embeddings. args | STRING |
| HIGH | lightx2v/models/schedulers/qwen_image/scheduler.py | 0 | this matches the implementation in denoising diffusion probabilistic models: create sinusoidal timestep embeddings. args | STRING |
| HIGH | lightx2v/models/schedulers/z_image/scheduler.py | 0 | this matches the implementation in denoising diffusion probabilistic models: create sinusoidal timestep embeddings. args | STRING |
| HIGH | lightx2v/models/video_encoders/hf/wan/vae_2_2.py | 0 | fix bfloat16 support for nearest neighbor interpolation. | STRING |
| HIGH | lightx2v/models/video_encoders/hf/wan/vae.py | 0 | fix bfloat16 support for nearest neighbor interpolation. | STRING |
| HIGH | …ain/lightx2v_train/model_zoo/native/wan/modules/vae.py | 0 | fix bfloat16 support for nearest neighbor interpolation. | STRING |
| HIGH | lightx2v/models/networks/bagel/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 | lightx2v/models/networks/bagel/modeling_utils.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 | lightx2v/models/networks/neopp/infer/pre_infer.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 | lightx2v/models/networks/worldplay/model.py | 0 | initialize inference modules and connect action weights. | STRING |
| HIGH | lightx2v/models/networks/worldplay/ar_model.py | 0 | initialize inference modules and connect action weights. | STRING |
| HIGH | lightx2v/models/networks/worldplay/bi_model.py | 0 | initialize inference modules and connect action weights. | STRING |
| HIGH | lightx2v/models/networks/worldplay/model.py | 0 | run inference with action and camera pose conditioning. args: inputs: dict containing: - text_encoder_output: text encod | STRING |
| HIGH | lightx2v/models/networks/worldplay/ar_model.py | 0 | run inference with action and camera pose conditioning. args: inputs: dict containing: - text_encoder_output: text encod | STRING |
| HIGH | lightx2v/models/networks/worldplay/bi_model.py | 0 | run inference with action and camera pose conditioning. args: inputs: dict containing: - text_encoder_output: text encod | STRING |
| HIGH | …odels/networks/worldplay/infer/ar_transformer_infer.py | 0 | modulate with per-token scale and shift (no unsqueeze). args: x: [l, c] tensor scale: [b, l, c] or [b, c] tensor shift: | STRING |
| HIGH | …v/models/networks/worldplay/infer/transformer_infer.py | 0 | modulate with per-token scale and shift (no unsqueeze). args: x: [l, c] tensor scale: [b, l, c] or [b, c] tensor shift: | STRING |
| HIGH | …odels/networks/worldplay/infer/bi_transformer_infer.py | 0 | modulate with per-token scale and shift (no unsqueeze). args: x: [l, c] tensor scale: [b, l, c] or [b, c] tensor shift: | STRING |
| HIGH | …odels/networks/worldplay/infer/ar_transformer_infer.py | 0 | apply projective positional encoding to q, k, v. args: q: query tensor [b, l, h, d] or [b, h, l, d] k: key tensor v: val | STRING |
| HIGH | …v/models/networks/worldplay/infer/transformer_infer.py | 0 | apply projective positional encoding to q, k, v. args: q: query tensor [b, l, h, d] or [b, h, l, d] k: key tensor v: val | STRING |
| HIGH | …odels/networks/worldplay/infer/bi_transformer_infer.py | 0 | apply projective positional encoding to q, k, v. args: q: query tensor [b, l, h, d] or [b, h, l, d] k: key tensor v: val | STRING |
| HIGH | …models/schedulers/hidream_o1_image/fm_solvers_unipc.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …ghtx2v/models/schedulers/hunyuan3d/flow_match_euler.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/fm_solvers_unipc.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …odel_zoo/native/lingbot_video/scheduling_flow_unipc.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …ightx2v_train/model_zoo/native/wan/utils/fm_solvers.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/fm_solvers_unipc.py | 0 | the index counter for current timestep. it will increase 1 after each scheduler step. | STRING |
| HIGH | …models/schedulers/hidream_o1_image/fm_solvers_unipc.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …ghtx2v/models/schedulers/hunyuan3d/flow_match_euler.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/fm_solvers_unipc.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …odel_zoo/native/lingbot_video/scheduling_flow_unipc.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …ightx2v_train/model_zoo/native/wan/utils/fm_solvers.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/fm_solvers_unipc.py | 0 | the index for the first timestep. it should be set from pipeline with `set_begin_index` method. | STRING |
| HIGH | …models/schedulers/hidream_o1_image/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 | …ghtx2v/models/schedulers/hunyuan3d/flow_match_euler.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/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 | …odel_zoo/native/lingbot_video/scheduling_flow_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 | …ightx2v_train/model_zoo/native/wan/utils/fm_solvers.py | 0 | sets the begin index for the scheduler. this function should be run from pipeline before the inference. args: begin_inde | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/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 | …models/schedulers/hidream_o1_image/fm_solvers_unipc.py | 0 | sets the discrete timesteps used for the diffusion chain (to be run before inference). args: num_inference_steps (`int`) | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/fm_solvers_unipc.py | 0 | sets the discrete timesteps used for the diffusion chain (to be run before inference). args: num_inference_steps (`int`) | STRING |
| HIGH | …odel_zoo/native/lingbot_video/scheduling_flow_unipc.py | 0 | sets the discrete timesteps used for the diffusion chain (to be run before inference). args: num_inference_steps (`int`) | STRING |
| HIGH | …ightx2v_train/model_zoo/native/wan/utils/fm_solvers.py | 0 | sets the discrete timesteps used for the diffusion chain (to be run before inference). args: num_inference_steps (`int`) | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/fm_solvers_unipc.py | 0 | sets the discrete timesteps used for the diffusion chain (to be run before inference). args: num_inference_steps (`int`) | STRING |
| HIGH | …models/schedulers/hidream_o1_image/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 | lightx2v/models/schedulers/wan/s2v/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 | …odel_zoo/native/lingbot_video/scheduling_flow_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 | …ightx2v_train/model_zoo/native/wan/utils/fm_solvers.py | 0 | "dynamic thresholding: at each sampling step we set s to a certain percentile absolute pixel value in xt0 (the predictio | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/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 | …models/schedulers/hidream_o1_image/fm_solvers_unipc.py | 0 | convert the model output to the corresponding type the unipc algorithm needs. args: model_output (`torch.tensor`): the d | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/fm_solvers_unipc.py | 0 | convert the model output to the corresponding type the unipc algorithm needs. args: model_output (`torch.tensor`): the d | STRING |
| HIGH | …odel_zoo/native/lingbot_video/scheduling_flow_unipc.py | 0 | convert the model output to the corresponding type the unipc algorithm needs. args: model_output (`torch.tensor`): the d | STRING |
| HIGH | …v_train/model_zoo/native/wan/utils/fm_solvers_unipc.py | 0 | convert the model output to the corresponding type the unipc algorithm needs. args: model_output (`torch.tensor`): the d | STRING |
| HIGH | …models/schedulers/hidream_o1_image/fm_solvers_unipc.py | 0 | one step for the unip (b(h) version). alternatively, `self.solver_p` is used if is specified. args: model_output (`torch | STRING |
| HIGH | lightx2v/models/schedulers/wan/s2v/fm_solvers_unipc.py | 0 | one step for the unip (b(h) version). alternatively, `self.solver_p` is used if is specified. args: model_output (`torch | STRING |
| HIGH | …odel_zoo/native/lingbot_video/scheduling_flow_unipc.py | 0 | one step for the unip (b(h) version). alternatively, `self.solver_p` is used if is specified. args: model_output (`torch | STRING |
| 47 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/preprocess/pose2d_utils.py | 1052 | except Exception as e: | CODE |
| LOW | tools/preprocess/ltx2.3_preprocess_data.py | 271 | except Exception: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1232 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 1263 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 2170 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 2189 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/sp_bench.py | 2697 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/operator_backends.py | 996 | except Exception as exc: | CODE |
| LOW | tools/benchmarks/operator_bench.py | 451 | except Exception as exc: | CODE |
| LOW | tools/extract/lora_merger.py | 412 | except Exception as e: | CODE |
| MEDIUM | tools/extract/lora_merger.py | 413 | print(f"Error: {e}") | CODE |
| LOW | tools/extract/convert_lightning_to_x2v_lora.py | 267 | except Exception as e: | CODE |
| MEDIUM | tools/extract/convert_lightning_to_x2v_lora.py | 268 | print(f"Error: {e}") | CODE |
| LOW | tools/extract/convert_vigen_to_x2v_model.py | 176 | except Exception as e: | CODE |
| MEDIUM | tools/extract/convert_vigen_to_x2v_model.py | 177 | print(f"Error: {e}") | CODE |
| LOW⚡ | tools/extract/lora_extractor.py | 372 | except Exception as e: | CODE |
| MEDIUM⚡ | tools/extract/lora_extractor.py | 373 | print(f"Error: {e}") | CODE |
| MEDIUM⚡ | tools/extract/lora_extractor.py | 377 | print(f"Error: {key} is not a 2D weight tensor") | CODE |
| LOW | tools/extract/lora_extractor.py | 450 | except Exception as e: | CODE |
| MEDIUM | tools/extract/lora_extractor.py | 451 | print(f"Error: {e}") | CODE |
| LOW | tools/postprocess/torchvision_fix.py | 68 | except Exception: | CODE |
| LOW | tools/postprocess/paint_pipeline.py | 249 | except Exception as exc: | CODE |
| LOW | tools/convert/seko_talk_converter.py | 408 | except Exception as e: | CODE |
| LOW | tools/convert/converter.py | 947 | except Exception as e: | CODE |
| LOW | tools/convert/converter.py | 1034 | except Exception as e: | CODE |
| LOW | lightx2v_platform/set_ai_device.py | 28 | except Exception: # pragma: no cover - importlib.metadata is stdlib on 3.8+ | CODE |
| LOW | lightx2v_platform/set_ai_device.py | 43 | except Exception as e: | CODE |
| LOW | lightx2v_platform/set_ai_device.py | 67 | except Exception as error: | CODE |
| LOW | …2v_platform/ops/rope/ascend_npu/minimax_h3_npu_rope.py | 32 | except Exception as exc: | CODE |
| MEDIUM | …2v_platform/ops/rope/ascend_npu/minimax_h3_npu_rope.py | 24 | def _load_mindiesd_rope(): | CODE |
| LOW | lightx2v_platform/ops/utils/biren_supa/sbp_utils.py | 12 | except Exception: | CODE |
| LOW | lightx2v_platform/ops/utils/biren_supa/sbp_utils.py | 31 | except Exception: | CODE |
| MEDIUM | lightx2v_platform/ops/utils/biren_supa/sbp_utils.py | 16 | def _tensor_sbp_axis(input_tensor): | CODE |
| LOW | lightx2v_platform/ops/attn/amd_rocm/sage_attn.py | 90 | except Exception: | CODE |
| LOW | lightx2v_platform/ops/attn/amd_rocm/sage_attn.py | 108 | except Exception as e: | CODE |
| LOW | lightx2v_platform/ops/attn/amd_rocm/sage_attn.py | 143 | except Exception: | CODE |
| LOW | lightx2v_platform/ops/mm/amd_rocm/mm_weight.py | 19 | except Exception: | CODE |
| MEDIUM | lightx2v_platform/ops/mm/amd_rocm/mm_weight.py | 16 | def _gcn_arch(): | CODE |
| LOW | lightx2v_platform/compilation/ascend_npu/mindie.py | 11 | except Exception as exc: | CODE |
| LOW | lightx2v_platform/base/biren_supa.py | 72 | except Exception: | CODE |
| MEDIUM | lightx2v_platform/base/biren_supa.py | 69 | def is_available() -> bool: | CODE |
| LOW | lightx2v_platform/base/enflame_gcu.py | 94 | except Exception: | CODE |
| LOW | lightx2v/server/task_manager.py | 325 | except Exception as e: | CODE |
| LOW | lightx2v/server/main.py | 83 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/files.py | 69 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/files.py | 84 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/deps.py | 57 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/server.py | 50 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/server.py | 154 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/openai_images.py | 139 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/common.py | 74 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/common.py | 133 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/common.py | 149 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/common.py | 163 | except Exception as e: | CODE |
| MEDIUM | lightx2v/server/api/tasks/common.py | 155 | def stop_all_running_tasks(): | CODE |
| LOW | lightx2v/server/api/tasks/video.py | 31 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/image.py | 112 | except Exception as e: | CODE |
| LOW | lightx2v/server/api/tasks/image.py | 174 | except Exception as e: | CODE |
| LOW | lightx2v/server/services/file_service.py | 72 | except Exception as e: | CODE |
| LOW | lightx2v/server/services/file_service.py | 119 | except Exception as e: | CODE |
| 352 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/preprocess/ltx2.3_preprocess_data.py | 271 | except Exception: | CODE |
| HIGH | tools/postprocess/torchvision_fix.py | 62 | except ImportError: | CODE |
| HIGH | tools/convert/quant/quant.py | 15 | except ImportError: | CODE |
| HIGH | lightx2v_platform/ops/utils/biren_supa/sbp_utils.py | 31 | except Exception: | CODE |
| HIGH | lightx2v_platform/ops/attn/amd_rocm/sage_attn.py | 143 | except Exception: | CODE |
| HIGH | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 177 | except RuntimeError: | CODE |
| HIGH | lightx2v_platform/base/enflame_gcu.py | 7 | except ImportError: | CODE |
| HIGH | lightx2v/server/main.py | 81 | except KeyboardInterrupt: | CODE |
| HIGH | lightx2v/server/services/distributed_utils.py | 73 | except Exception: | CODE |
| HIGH | lightx2v/utils/va_recorder.py | 541 | except OSError: | CODE |
| HIGH | lightx2v/utils/va_recorder.py | 550 | except OSError: | CODE |
| HIGH | lightx2v/utils/video_recorder.py | 312 | except OSError: | CODE |
| HIGH | lightx2v/utils/profiler.py | 125 | except (RuntimeError, NotImplementedError): | CODE |
| HIGH | lightx2v/models/networks/worldmirror/model.py | 854 | except Exception: | CODE |
| HIGH | …htx2v/models/networks/worldmirror/utils/video_utils.py | 526 | except Exception: | CODE |
| HIGH | …htx2v/models/networks/worldmirror/utils/visual_util.py | 260 | except (ValueError, IndexError): | CODE |
| HIGH | …htx2v/models/networks/neopp/infer/transformer_infer.py | 72 | except ImportError: | CODE |
| HIGH | …x2v/models/networks/hunyuan3d/utils/torchvision_fix.py | 77 | except ImportError: | CODE |
| HIGH | lightx2v/models/networks/seedvr/model.py | 146 | except Exception: | CODE |
| HIGH | lightx2v/models/runners/base_runner.py | 350 | except Exception: | CODE |
| HIGH | lightx2v/models/runners/base_runner.py | 357 | except Exception: | CODE |
| HIGH | lightx2v/models/runners/base_runner.py | 366 | except Exception: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_matrix_game2_runner.py | 93 | except Exception as e: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_matrix_game2_runner.py | 121 | except Exception as e: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_matrix_game2_runner.py | 145 | except Exception as e: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_animate2_runner.py | 135 | except OSError: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_animate2_runner.py | 139 | except OSError: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_animate2_runner.py | 154 | except OSError: | CODE |
| HIGH | lightx2v/models/runners/wan/wan_infinitetalk_runner.py | 332 | except OSError: | CODE |
| HIGH | lightx2v/disagg/rdma_server.py | 154 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_server.py | 186 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1219 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1223 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1228 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1237 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1243 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1254 | except Exception: | CODE |
| HIGH | lightx2v/disagg/disagg_mixin.py | 1262 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_client.py | 121 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_client.py | 146 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_client.py | 195 | except Exception: | CODE |
| HIGH | lightx2v/disagg/mooncake.py | 21 | except Exception: | CODE |
| HIGH | lightx2v/disagg/mooncake.py | 28 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 22 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 34 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 41 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 59 | except ValueError: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 207 | except Exception: | CODE |
| HIGH | lightx2v/disagg/rdma_utils.py | 110 | except Exception: | CODE |
| HIGH | lightx2v/disagg/examples/run_controller.py | 190 | except OSError: | CODE |
| HIGH | lightx2v/disagg/services/controller.py | 117 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/controller.py | 141 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/controller.py | 716 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/controller.py | 1072 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/encoder.py | 260 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/encoder.py | 317 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/transformer.py | 288 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/transformer.py | 331 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/transformer.py | 915 | except Exception: | CODE |
| HIGH | lightx2v/disagg/services/transformer.py | 484 | except Exception: | CODE |
| 27 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/cache_minimax_h3_adaln/builder.py | 104 | CODE | |
| LOW | tools/preprocess/pose2d.py | 205 | CODE | |
| LOW | tools/preprocess/pose2d.py | 347 | CODE | |
| LOW | tools/preprocess/retarget_pose.py | 153 | CODE | |
| LOW | tools/preprocess/retarget_pose.py | 357 | CODE | |
| LOW | tools/preprocess/retarget_pose.py | 765 | CODE | |
| LOW | tools/preprocess/process_pipepline.py | 159 | CODE | |
| LOW | tools/preprocess/process_pipepline.py | 435 | CODE | |
| LOW | tools/preprocess/pose2d_utils.py | 10 | CODE | |
| LOW | tools/preprocess/pose2d_utils.py | 773 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 1181 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 1940 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 2071 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 2327 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 2515 | CODE | |
| LOW | tools/benchmarks/sp_bench.py | 2613 | CODE | |
| LOW | tools/benchmarks/operator_backends.py | 255 | CODE | |
| LOW | tools/benchmarks/operator_bench.py | 476 | CODE | |
| LOW | tools/benchmarks/operator_bench.py | 648 | CODE | |
| LOW | tools/extract/lora_merger.py | 66 | CODE | |
| LOW | tools/extract/lora_merger.py | 199 | CODE | |
| LOW | tools/extract/convert_lightning_to_x2v_lora.py | 33 | CODE | |
| LOW | tools/extract/convert_lightning_to_x2v_lora.py | 91 | CODE | |
| LOW | tools/extract/lora_extractor.py | 68 | CODE | |
| LOW | tools/extract/lora_extractor.py | 287 | CODE | |
| LOW | tools/postprocess/torchvision_fix.py | 7 | CODE | |
| LOW | tools/postprocess/paint_pipeline.py | 131 | CODE | |
| LOW | tools/convert/seko_talk_converter.py | 334 | CODE | |
| LOW | tools/convert/converter.py | 336 | CODE | |
| LOW | tools/convert/converter.py | 618 | CODE | |
| LOW | tools/convert/converter.py | 738 | CODE | |
| LOW | tools/convert/converter.py | 1018 | CODE | |
| LOW | lightx2v_platform/ops/rope/ascend_npu/npu_rope.py | 46 | CODE | |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 45 | CODE | |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 198 | CODE | |
| LOW | lightx2v/pipeline.py | 18 | CODE | |
| LOW | lightx2v/pipeline.py | 250 | CODE | |
| LOW | lightx2v/pipeline.py | 301 | CODE | |
| LOW | lightx2v/shot_runner/utils.py | 175 | CODE | |
| LOW | lightx2v/server/metrics/metrics.py | 379 | CODE | |
| LOW | lightx2v/server/api/openai_images.py | 54 | CODE | |
| LOW | lightx2v/server/services/file_service.py | 45 | CODE | |
| LOW | lightx2v/server/services/inference/worker.py | 73 | CODE | |
| LOW | lightx2v/server/services/inference/worker.py | 144 | CODE | |
| LOW | lightx2v/server/services/inference/worker.py | 185 | CODE | |
| LOW | lightx2v/server/services/generation/image.py | 17 | CODE | |
| LOW | lightx2v/server/media/audio.py | 33 | CODE | |
| LOW | lightx2v/server/media/image.py | 31 | CODE | |
| LOW | lightx2v/server/media/base.py | 33 | CODE | |
| LOW | lightx2v/utils/set_config.py | 69 | CODE | |
| LOW | lightx2v/utils/va_reader_omni.py | 205 | CODE | |
| LOW | lightx2v/utils/va_reader_omni.py | 476 | CODE | |
| LOW | lightx2v/utils/va_reader_omni.py | 766 | CODE | |
| LOW | lightx2v/utils/va_recorder.py | 87 | CODE | |
| LOW | lightx2v/utils/va_recorder.py | 123 | CODE | |
| LOW | lightx2v/utils/va_recorder.py | 467 | CODE | |
| LOW | lightx2v/utils/va_recorder_x264.py | 57 | CODE | |
| LOW | lightx2v/utils/va_recorder_x264.py | 177 | CODE | |
| LOW | lightx2v/utils/ggml_tensor.py | 341 | CODE | |
| LOW | lightx2v/utils/utils.py | 218 | CODE | |
| 327 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/preprocess/pose2d.py | 56 | CODE | |
| LOW | tools/preprocess/human_visualization.py | 255 | CODE | |
| LOW | tools/preprocess/human_visualization.py | 575 | CODE | |
| LOW | tools/preprocess/human_visualization.py | 728 | CODE | |
| LOW | tools/preprocess/retarget_pose.py | 762 | CODE | |
| LOW | tools/preprocess/process_pipepline.py | 159 | CODE | |
| LOW | tools/benchmarks/operator_bench.py | 476 | CODE | |
| LOW | tools/postprocess/paint_pipeline.py | 200 | CODE | |
| LOW | tools/postprocess/paint_pipeline.py | 231 | CODE | |
| LOW | tools/convert/converter.py | 336 | CODE | |
| LOW | lightx2v_platform/ops/norm/norm_template.py | 42 | CODE | |
| LOW | lightx2v_platform/ops/norm/norm_template.py | 171 | CODE | |
| LOW | lightx2v_platform/ops/norm/biren_supa/br_rms_norm.py | 14 | CODE | |
| LOW | lightx2v_platform/ops/norm/biren_supa/supa_rms_norm.py | 10 | CODE | |
| LOW | lightx2v_platform/ops/norm/metax_cuda/metax_rms_norm.py | 51 | CODE | |
| LOW | lightx2v_platform/ops/norm/ascend_npu/npu_layer_norm.py | 14 | CODE | |
| LOW | lightx2v_platform/ops/norm/ascend_npu/npu_layer_norm.py | 47 | CODE | |
| LOW | …ghtx2v_platform/ops/norm/cambricon_mlu/mlu_rms_norm.py | 12 | CODE | |
| LOW | …v_platform/ops/norm/iluvatar_cuda/iluvatar_rms_norm.py | 18 | CODE | |
| LOW | lightx2v_platform/ops/norm/enflame_gcu/layer_norm.py | 9 | CODE | |
| LOW | lightx2v_platform/ops/attn/biren_supa/flash_attn.py | 96 | CODE | |
| LOW | lightx2v_platform/ops/attn/biren_supa/supa_sdpa.py | 15 | CODE | |
| LOW | lightx2v_platform/ops/attn/biren_supa/ulysses_attn.py | 39 | CODE | |
| LOW | lightx2v_platform/ops/attn/biren_supa/ulysses_attn.py | 137 | CODE | |
| LOW | lightx2v_platform/ops/attn/hygon_dcu/flash_attn.py | 52 | CODE | |
| LOW | lightx2v_platform/ops/attn/hygon_dcu/flash_attn.py | 183 | CODE | |
| LOW | lightx2v_platform/ops/attn/enflame_gcu/flash_attn.py | 60 | CODE | |
| LOW | lightx2v_platform/ops/attn/enflame_gcu/flash_attn.py | 247 | CODE | |
| LOW | lightx2v_platform/ops/mm/biren_supa/mm_weight.py | 74 | CODE | |
| LOW | lightx2v_platform/ops/mm/mthreads_musa/mm_weight.py | 17 | CODE | |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 314 | CODE | |
| LOW | lightx2v_platform/ops/mm/intel_xpu/mm_weight.py | 404 | CODE | |
| LOW | lightx2v_platform/ops/mm/amd_rocm/mm_weight.py | 103 | CODE | |
| LOW | lightx2v/pipeline.py | 18 | CODE | |
| LOW | lightx2v/pipeline.py | 110 | CODE | |
| LOW | lightx2v/pipeline.py | 250 | CODE | |
| LOW | lightx2v/pipeline.py | 355 | CODE | |
| LOW | lightx2v/pipeline.py | 390 | CODE | |
| LOW | lightx2v/training/worldplay_ar_pipeline.py | 246 | CODE | |
| LOW | lightx2v/server/api/openai_images.py | 252 | CODE | |
| LOW | lightx2v/server/api/tasks/video.py | 37 | CODE | |
| LOW | lightx2v/server/api/tasks/image.py | 180 | CODE | |
| LOW | lightx2v/utils/va_reader_omni.py | 295 | CODE | |
| LOW | lightx2v/utils/va_reader_omni.py | 563 | CODE | |
| LOW | lightx2v/utils/torch_trace_profiler.py | 180 | CODE | |
| LOW | lightx2v/utils/ggml_tensor.py | 18 | CODE | |
| LOW | …htx2v/models/video_encoders/hf/ltx2/upsampler/model.py | 35 | CODE | |
| LOW | …x2v/models/video_encoders/hf/ltx2/audio_vae/vocoder.py | 324 | CODE | |
| LOW | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 142 | CODE | |
| LOW | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 362 | CODE | |
| LOW | …models/video_encoders/hf/ltx2/video_vae/convolution.py | 11 | CODE | |
| LOW | …v/models/video_encoders/hf/ltx2/video_vae/video_vae.py | 268 | CODE | |
| LOW | …v/models/video_encoders/hf/ltx2/video_vae/video_vae.py | 522 | CODE | |
| LOW | …_encoders/hf/ltx2/video_vae/diffusion_video_decoder.py | 605 | CODE | |
| LOW | …tx2v/models/video_encoders/hf/ltx2/video_vae/resnet.py | 24 | CODE | |
| LOW | …tx2v/models/video_encoders/hf/ltx2/video_vae/resnet.py | 200 | CODE | |
| LOW | lightx2v/models/video_encoders/hf/lingbot_video/vae.py | 54 | CODE | |
| LOW | lightx2v/models/video_encoders/hf/lingbot_video/vae.py | 127 | CODE | |
| LOW | …ders/hf/hunyuan3d/autoencoders/attention_processors.py | 32 | CODE | |
| LOW | …encoders/hf/hunyuan3d/autoencoders/attention_blocks.py | 415 | CODE | |
| 170 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | lightx2v_platform/ops/a2a/ascend_npu/__init__.py | 3 | __all__ = ["HcclEagerUlyssesA2A"] | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/__init__.py | 3 | __all__ = ["MetaxRope"] | CODE |
| LOW | lightx2v_platform/ops/rope/metax_cuda/metax_rope.py | 227 | __all__ = ["MetaxRope"] | CODE |
| LOW | lightx2v_platform/ops/rope/ascend_npu/__init__.py | 4 | __all__ = ["NpuRope", "MiniMaxH3NpuRope"] | CODE |
| LOW | lightx2v_platform/ops/rope/cambricon_mlu/__init__.py | 3 | __all__ = ["MluRope"] | CODE |
| LOW | lightx2v_platform/ops/rope/iluvatar_cuda/__init__.py | 3 | __all__ = ["IluvatarWanRope"] | CODE |
| LOW | lightx2v_platform/ops/rope/enflame_gcu/__init__.py | 3 | __all__ = ["GcuWanRope"] | CODE |
| LOW | lightx2v_platform/ops/utils/biren_supa/__init__.py | 3 | __all__ = ["convBB", "convSB"] | CODE |
| LOW | lightx2v_platform/ops/norm/metax_cuda/__init__.py | 3 | __all__ = ["MetaxRmsNormWeight"] | CODE |
| LOW | lightx2v_platform/ops/norm/metax_cuda/metax_rms_norm.py | 108 | __all__ = ["MetaxRmsNormWeight"] | CODE |
| LOW | lightx2v_platform/ops/norm/ascend_npu/__init__.py | 4 | __all__ = ["NpuRmsNormWeight", "NpuLayerNormWeight"] | CODE |
| LOW | lightx2v_platform/ops/norm/iluvatar_cuda/__init__.py | 3 | __all__ = ["IluvatarRmsNormWeight"] | CODE |
| LOW | lightx2v_platform/ops/norm/enflame_gcu/__init__.py | 3 | __all__ = ["GcuLayerNormWeight"] | CODE |
| LOW | …htx2v_platform/ops/attn/metax_cuda/metax_flash_attn.py | 180 | __all__ = ["MetaxFlashAttn2Weight"] | CODE |
| LOW | lightx2v_platform/ops/attn/metax_cuda/__init__.py | 4 | __all__ = ["MetaxFlashAttn2Weight", "MetaxSageAttn2Weight"] | CODE |
| LOW | lightx2v_platform/ops/mm/mthreads_musa/__init__.py | 4 | __all__ = [ | CODE |
| LOW | lightx2v_platform/ops/mm/mthreads_musa/fp8_scaled_mm.py | 188 | __all__ = ["FP8_DTYPE", "fp8_linear", "fp8_scaled_mm", "per_token_quant_fp8"] | CODE |
| LOW | lightx2v_platform/ops/mm/mthreads_musa/mm_weight.py | 63 | __all__ = ["MMWeightWfp8channelAfp8tokendynamicMusa"] | CODE |
| LOW | lightx2v_platform/ops/moe/__init__.py | 3 | __all__ = ["FusedMoEActivation", "FusedMoETemplate"] | CODE |
| LOW | lightx2v_platform/ops/moe/template.py | 143 | __all__ = ["FusedMoEActivation", "FusedMoETemplate"] | CODE |
| LOW | …htx2v_platform/ops/moe/metax_cuda/metax_mctlass_moe.py | 275 | __all__ = ["MetaxMctlassFusedMoE"] | CODE |
| LOW | lightx2v_platform/ops/moe/metax_cuda/__init__.py | 3 | __all__ = ["MetaxMctlassFusedMoE"] | CODE |
| LOW | lightx2v_platform/ops/moe/ascend_npu/__init__.py | 3 | __all__ = ["NpuGroupedMMFusedMoE"] | CODE |
| LOW | lightx2v_platform/ops/moe/ascend_npu/npu_grouped_mm.py | 169 | __all__ = ["NpuGroupedMMFusedMoE"] | CODE |
| LOW | lightx2v_platform/compilation/ascend_npu/__init__.py | 3 | __all__ = ["create_mindie_backend"] | CODE |
| LOW | lightx2v_platform/base/__init__.py | 16 | __all__ = [ | CODE |
| LOW | lightx2v_platform/base/amd_rocm.py | 158 | __all__ = [ | CODE |
| LOW | lightx2v/__init__.py | 9 | __all__ = [ | CODE |
| LOW | lightx2v/training/__init__.py | 10 | __all__ = [ | CODE |
| LOW | lightx2v/server/api/__init__.py | 4 | __all__ = [ | CODE |
| LOW | lightx2v/server/api/tasks/__init__.py | 5 | __all__ = [ | CODE |
| LOW | lightx2v/server/services/__init__.py | 5 | __all__ = [ | CODE |
| LOW | lightx2v/server/services/inference/__init__.py | 4 | __all__ = [ | CODE |
| LOW | lightx2v/server/services/generation/__init__.py | 5 | __all__ = [ | CODE |
| LOW | lightx2v/server/media/__init__.py | 5 | __all__ = [ | CODE |
| LOW | lightx2v/utils/ltx2_media_io.py | 24 | logger = logging.getLogger(__name__) | CODE |
| LOW | …2v/models/video_encoders/hf/ltx2/audio_vae/__init__.py | 15 | __all__ = [ | CODE |
| LOW | …v/models/video_encoders/hf/ltx2/video_vae/triton_na.py | 168 | __all__ = ["na3d"] | CODE |
| LOW | …2v/models/video_encoders/hf/ltx2/video_vae/__init__.py | 15 | __all__ = [ | CODE |
| LOW | …_encoders/hf/ltx2/video_vae/diffusion_video_decoder.py | 30 | logger = logging.getLogger(__name__) | CODE |
| LOW | …_encoders/hf/ltx2/video_vae/diffusion_video_decoder.py | 998 | __all__ = ["DiffusionVideoDecoder"] | CODE |
| LOW | …dels/video_encoders/hf/ltx2/video_vae/triton_swiglu.py | 233 | __all__ = ["residual_modulating_mlp", "swiglu_tiled", "triton_swiglu_available"] | CODE |
| LOW | …s/video_encoders/hf/ltx2/video_vae/diffusion_tiling.py | 189 | __all__ = [ | CODE |
| LOW | …htx2v/models/video_encoders/hf/minimax_h3/video_vae.py | 903 | def set_decode_tile_shape(self, tile_height: int, tile_width: int) -> None: | CODE |
| LOW | lightx2v/models/video_encoders/hf/wan/vae_2_2.py | 16 | __all__ = [ | CODE |
| LOW | lightx2v/models/video_encoders/hf/wan/vae.py | 17 | __all__ = [ | CODE |
| LOW | …x2v/models/video_encoders/hf/longcat_image/__init__.py | 3 | __all__ = ["LongCatImageVAE"] | CODE |
| LOW | lightx2v/models/video_encoders/hf/seedvr/__init__.py | 3 | __all__ = ["attn_video_vae_v3_s8_c16_t4_inflation_sd3_init"] | CODE |
| LOW⚡ | …ideo_encoders/hf/seedvr/common/distributed/advanced.py | 77 | def set_sequence_parallel_group(group: Optional[dist.ProcessGroup]) -> None: | CODE |
| LOW | …els/video_encoders/hf/seedvr/common/distributed/ops.py | 29 | def set_sequence_parallel_a2a_backend(backend: str) -> None: | CODE |
| LOW | …odels/networks/qwen_image_21/infer/offload/__init__.py | 3 | __all__ = ["QwenImage21OffloadTransformerInfer"] | CODE |
| LOW | lightx2v/models/networks/hunyuan_image3/config.py | 273 | __all__ = [ | CODE |
| LOW | …2v/models/networks/hunyuan_image3/custom_all_reduce.py | 177 | __all__ = ["HunyuanImage3CustomAllReduce", "HunyuanImage3CustomAllReduceConfig"] | CODE |
| LOW | lightx2v/models/networks/hunyuan_image3/__init__.py | 1 | __all__ = ["HunyuanImage3Model"] | CODE |
| LOW | …2v/models/networks/hunyuan_image3/weights/hybrid_tp.py | 310 | __all__ = [ | CODE |
| LOW | …x2v/models/networks/hunyuan_image3/weights/__init__.py | 5 | __all__ = [ | CODE |
| LOW | …ghtx2v/models/networks/worldmirror/weights/__init__.py | 6 | __all__ = [ | CODE |
| LOW | …tworks/worldmirror/models/models/visual_transformer.py | 15 | logger = logging.getLogger(__name__) | CODE |
| LOW | lightx2v/models/networks/flux2/__init__.py | 3 | __all__ = ["Flux2KleinTransformerModel", "Flux2DevTransformerModel"] | CODE |
| LOW | lightx2v/models/networks/flux2/weights/__init__.py | 5 | __all__ = [ | CODE |
| 145 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | lightx2v/server/services/generation/base.py | 31 | task_type = self.inference_service.worker.runner.config.get("task", "t2v") | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 22 | dst.conv1_local.conv.weight.data.copy_(src.conv1_local_weight.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 23 | dst.conv1_local.conv.bias.data.copy_(src.conv1_local_bias.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 24 | dst.conv2.conv.weight.data.copy_(src.conv2_weight.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 25 | dst.conv2.conv.bias.data.copy_(src.conv2_bias.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 26 | dst.conv3.conv.weight.data.copy_(src.conv3_weight.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 27 | dst.conv3.conv.bias.data.copy_(src.conv3_bias.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 30 | dst.conv1_global.conv.weight.data.copy_(src.conv1_global_weight.tensor) | CODE |
| CRITICAL⚡ | …/models/networks/wan/infer/s2v/causal_audio_encoder.py | 31 | dst.conv1_global.conv.bias.data.copy_(src.conv1_global_bias.tensor) | CODE |
| CRITICAL | lightx2v/models/runners/flux2/flux2_runner.py | 458 | bn_mean = self.vae.vae.bn.running_mean.view(1, -1, 1, 1).to(latents.device, latents.dtype) | CODE |
| CRITICAL | lightx2v/models/runners/flux2/flux2_runner.py | 459 | bn_std = torch.sqrt(self.vae.vae.bn.running_var.view(1, -1, 1, 1) + self.vae.vae.config.batch_norm_eps).to(laten | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …s/cache_minimax_h3_adaln/run_cache_minimax_h3_adaln.sh | 21 | # Apple MPS | COMMENT |
| LOW | …s/cache_minimax_h3_adaln/run_cache_minimax_h3_adaln.sh | 41 | # export PLATFORM=hygon_dcu | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 301 | # kp2ds_body = kp2ds_body[:18] | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 321 | [15, 17], | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 461 | kp2ds = kp2ds.copy() | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 481 | # [12, 13], | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 501 | # [0, 255, 170], | COMMENT |
| LOW | tools/preprocess/human_visualization.py | 521 | # keypoint2 = kp2ds_body[k2_index - 1] | COMMENT |
| LOW | tools/preprocess/ltx2.3_preprocess_data.py | 1 | # Copyright 2024-2025 The Alibaba Wan Team Authors. All rights reserved. | COMMENT |
| LOW | tools/convert/h3_video_vae_encoder.py | 1 | # Copyright 2026 The LightX2V Team. | COMMENT |
| LOW | lightx2v/utils/va_reader_omni.py | 601 | COMMENT | |
| LOW | lightx2v/utils/va_reader_omni.py | 621 | # ... | COMMENT |
| LOW | …2v/models/video_encoders/hf/ltx2/audio_vae/upsample.py | 21 | self.causality_axis = causality_axis | COMMENT |
| LOW | …ders/hf/hunyuan3d/autoencoders/attention_processors.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …s/video_encoders/hf/hunyuan3d/autoencoders/__init__.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …encoders/hf/hunyuan3d/autoencoders/attention_blocks.py | 1 | # Open Source Model Licensed under the Apache License Version 2.0 | COMMENT |
| LOW | …dels/video_encoders/hf/hunyuan3d/autoencoders/model.py | 1 | # Open Source Model Licensed under the Apache License Version 2.0 | COMMENT |
| LOW | …coders/hf/hunyuan3d/autoencoders/surface_extractors.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …_encoders/hf/hunyuan3d/autoencoders/volume_decoders.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …htx2v/models/video_encoders/hf/minimax_h3/video_vae.py | 1 | # Copyright 2026 The MiniMax and HuggingFace Teams. All rights reserved. | COMMENT |
| LOW | …ideo_encoders/hf/minimax_h3/fp8_encoder_conv_policy.py | 1 | # Copyright 2026 The LightX2V Team. | COMMENT |
| LOW | lightx2v/models/video_encoders/hf/minimax_h3/weights.py | 1 | # Copyright 2026 The LightX2V Team. | COMMENT |
| LOW | lightx2v/models/networks/worldmirror/model.py | 221 | config has no concept of quantization. | COMMENT |
| LOW | lightx2v/models/networks/worldmirror/model.py | 261 | # per block per forward (cheap in practice for single-scene | COMMENT |
| LOW | lightx2v/models/networks/worldmirror/model.py | 461 | if src is not None and src.shape == tensor.shape: | COMMENT |
| LOW | …ls/networks/worldmirror/weights/transformer_weights.py | 241 | self.cam_trunk_depth = cam_trunk_depth | COMMENT |
| LOW | …ls/networks/worldmirror/weights/transformer_weights.py | 521 | # Quantized leaves own their storage dtype (fp8/int8/nvfp4/...). | COMMENT |
| LOW | lightx2v/models/networks/worldmirror/comm/padding.py | 21 | # 计算需要填充的长度 | COMMENT |
| LOW | lightx2v/models/networks/worldmirror/comm/padding.py | 41 | COMMENT | |
| LOW | …v/models/networks/worldmirror/models/utils/sh_utils.py | 1 | # Copyright 2021 The PlenOctree Authors. | COMMENT |
| LOW | …models/networks/worldmirror/models/heads/dense_head.py | 141 | COMMENT | |
| LOW | lightx2v/models/networks/bagel/modeling_utils.py | 1 | # Copyright (c) 2022 Facebook, Inc. and its affiliates. | COMMENT |
| LOW | lightx2v/models/networks/worldplay/prope/camera_rope.py | 1 | # MIT License | COMMENT |
| LOW | …x2v/models/networks/hunyuan3d/utils/hy3dshape/utils.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …/models/input_encoders/hf/hunyuan3d/image_processor.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …tx2v/models/input_encoders/hf/hunyuan3d/conditioner.py | 1 | # Open Source Model Licensed under the Apache License Version 2.0 | COMMENT |
| LOW | lightx2v/models/input_encoders/hf/hunyuan3d/rembg.py | 1 | # Hunyuan 3D is licensed under the TENCENT HUNYUAN NON-COMMERCIAL LICENSE AGREEMENT | COMMENT |
| LOW | …htx2v/models/audio_encoders/hf/minimax_h3/audio_vae.py | 1 | # Copyright 2025 The MiniMax authors and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | …htx2v/models/runners/worldmirror/worldmirror_runner.py | 281 | # Build the wrapped model (nn.Module + side-car WeightModule) and | COMMENT |
| LOW | lightx2v/models/schedulers/neopp/scheduler.py | 61 | self.timesteps = self._apply_time_schedule(timesteps, timestep_shift=shift) | COMMENT |
| LOW | …ghtx2v/models/schedulers/hunyuan3d/flow_match_euler.py | 1 | # Copyright 2024 Stability AI, Katherine Crowson and The HuggingFace Team. All rights reserved. | COMMENT |
| LOW | lightx2v/disagg/rdma_server.py | 281 | # 单边写通常不需要接收方做额外操作，除非使用了带立即数或原子操作需要确认 | COMMENT |
| LOW | lightx2v/disagg/rdma_server.py | 301 | # conn = srv.handshake(host=args.listen_host, port=args.listen_port) | COMMENT |
| LOW | lightx2v/disagg/rdma_client.py | 461 | # # msg = b"Hello RDMA!" | COMMENT |
| LOW | lightx2v/disagg/rdma_client.py | 481 | # print(f"FAA old value: {old_value}") | COMMENT |
| LOW | lightx2v/common/ops/attn/svg2_attn_utils.py | 1161 | # --- End of loop over K blocks --- | COMMENT |
| LOW | lightx2v/common/ops/mm/mm_weight.py | 2581 | # just materialize ``weight`` and ``weight_scale`` here. | COMMENT |
| LOW | …d256/assets/rev256_onednn_v2_84tflops_pipelined_qk.cpp | 1 | #include <sycl/sycl.hpp> | COMMENT |
| LOW | …-hd256/assets/rev256_onednn_v2_83tflops_mad_fusion.cpp | 1 | #include <sycl/sycl.hpp> | COMMENT |
| LOW | …sdp-hd256/assets/rev256_onednn_v2_88tflops_s_scatter.h | 1 | // Approach 2: Eliminate S Register Transpose via lsc_slm_scatter Write + Block Read | COMMENT |
| LOW | …sdp-hd256/assets/rev256_onednn_v2_88tflops_s_scatter.h | 361 | // Four scatter calls per qp: kv groups 0..3, 4..7, 8..11, 12..15. | COMMENT |
| LOW | …hd256/assets/rev256_onednn_v2_86tflops_qk_pipeline.cpp | 1 | #include <sycl/sycl.hpp> | COMMENT |
| LOW | …p-hd256/assets/rev256_onednn_v2_88tflops_s_scatter.cpp | 1 | #include <sycl/sycl.hpp> | COMMENT |
| LOW | …dp-hd256/assets/rev256_onednn_v2_83tflops_mad_fusion.h | 1 | // oneDNN-inspired Flash Attention v2 kernel for HD=256 | COMMENT |
| LOW | …ls/xe2-sdp-hd256/assets/flash.attn.b.mha256.fp16.opt.h | 1 | // Optimized FP16 Flash Attention kernel for HD=256 | COMMENT |
| LOW | …p-hd256/assets/rev256_onednn_v2_86tflops_qk_pipeline.h | 1 | // oneDNN-inspired Flash Attention v2 kernel for HD=256 | COMMENT |
| LOW | …_skills/xe2-sdp-hd256/assets/test_rev256_onednn_v2.cpp | 1 | #include <sycl/sycl.hpp> | COMMENT |
| LOW | …-hd256/assets/rev256_onednn_v2_84tflops_pipelined_qk.h | 1 | // oneDNN-inspired Flash Attention v2 kernel for HD=256 | COMMENT |
| LOW | …d256/assets/rev256_onednn_v2_82tflops_fp16_transpose.h | 1 | // oneDNN-inspired Flash Attention v2 kernel for HD=256 | COMMENT |
| LOW | …skills/onednn-fp8-gemm/assets/fp16xfp8_per_n_scale.cpp | 1 | // ============================================================================= | COMMENT |
| 83 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 522 | # Step 1: Crop first to avoid exceeding target dimensions | COMMENT |
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 525 | # Step 2: Calculate padding needed for time and frequency dimensions | COMMENT |
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 529 | # Step 3: Apply padding if needed | COMMENT |
| LOW | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 541 | # Step 4: Final safety crop to ensure exact target shape | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 53 | # Step 1: RMSNorm | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 68 | # Step 2: Load scale and shift based on format | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 87 | # Step 3: Apply modulation: x_hat * (1 + scale) + shift | COMMENT |
| LOW | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 233 | # Step 1: Cache text KV (called once at generation start) | COMMENT |
| LOW⚡ | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 259 | # Step 2: For non-first chunks, cache context frame KV | COMMENT |
| LOW⚡ | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 264 | # Step 3: Run denoising for this chunk | COMMENT |
| LOW | …ic_skills/xe2-esimd-gemv/references/kernel-patterns.md | 109 | // Step 1: init SLM — MUST be first | COMMENT |
| LOW | …ic_skills/xe2-esimd-gemv/references/kernel-patterns.md | 122 | // Step 2: compute partial dot product | COMMENT |
| LOW⚡ | …ic_skills/xe2-esimd-gemv/references/kernel-patterns.md | 137 | // Step 3: store partial sum to SLM | COMMENT |
| LOW⚡ | …ic_skills/xe2-esimd-gemv/references/kernel-patterns.md | 144 | // Step 4: synchronize | COMMENT |
| LOW⚡ | …ic_skills/xe2-esimd-gemv/references/kernel-patterns.md | 147 | // Step 5: reduce and write output (k_thread_id == 0 only) | COMMENT |
| LOW⚡ | docs/EN/source/getting_started/quickstart.md | 68 | #### Step 1: Clone Repository | COMMENT |
| LOW⚡ | docs/EN/source/getting_started/quickstart.md | 76 | #### Step 2: Create Conda Virtual Environment | COMMENT |
| LOW⚡ | docs/EN/source/getting_started/quickstart.md | 84 | #### Step 3: Install Dependencies | COMMENT |
| LOW⚡ | docs/EN/source/getting_started/quickstart.md | 90 | #### Step 4: Install Attention Operators | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 109 | #### Step 4: Install Quantization Operators (Optional) | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 149 | #### Step 5: Verify Installation | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 162 | #### Step 1: Check CUDA Version | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 172 | #### Step 2: Create Python Environment | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 184 | #### Step 3: Install PyTorch Framework | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 211 | #### Step 4: Install Windows Version vLLM | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 225 | #### Step 5: Install Attention Mechanism Operators | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 246 | #### Step 6: Clone Repository | COMMENT |
| LOW | docs/EN/source/getting_started/quickstart.md | 258 | #### Step 7: Install Quantization Operators (Optional) | COMMENT |
| LOW | examples/BeginnerGuide/EN/02.Wan21-1.3B.md | 219 | # Step 1: Create LightX2VPipeline | COMMENT |
| LOW | examples/BeginnerGuide/EN/02.Wan21-1.3B.md | 226 | # Step 2: Set runtime parameters | COMMENT |
| LOW | examples/BeginnerGuide/EN/02.Wan21-1.3B.md | 246 | # Step 3: Start generating videos, can generate multiple times | COMMENT |
| LOW | scripts/server/post_multi_servers.py | 50 | # Step 2: Monitor task status until completion | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 52 | # Step 1: Send to Decoder first (sets up Phase2 receiver) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 59 | # Step 2: Send to Transformer (Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 66 | # Step 3: Send to Encoder (text encoding + Phase1 send) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 72 | # Step 4: Poll Decoder for completion (image saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 60 | # Step 1: Send to Decoder first (sets up Phase2 receiver) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 67 | # Step 2: Send to Transformer (Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 74 | # Step 3: Send to Encoder (text + image encoding + Phase1 send) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 80 | # Step 4: Poll Decoder for completion (image saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 46 | # Step 1: Send to Decoder first (sets up Phase2 receiver, starts blocking) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 53 | # Step 2: Send to Transformer (sets up Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 60 | # Step 3: Send to Encoder (triggers T5 encoding + Phase1 Mooncake send) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 66 | # Step 4: Poll Decoder for final completion (video is saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 57 | # Step 1: Send to Decoder first (sets up Phase2 receiver, starts blocking) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 64 | # Step 2: Send to Transformer (sets up Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 71 | # Step 3: Send to Encoder (triggers T5/CLIP/VAE encoding + Phase1 Mooncake send) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 77 | # Step 4: Poll Decoder for final completion (video is saved on Decoder node) | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …/runners/minimax_h3_causal/minimax_h3_causal_runner.py | 254 | for frames in decoder.push(video): | CODE |
| HIGH⚡ | lightx2v/disagg/services/controller.py | 657 | " kill -TERM ${pid} >/dev/null 2>&1 || true", | CODE |
| HIGH⚡ | lightx2v/disagg/services/controller.py | 661 | " kill -KILL ${pid} >/dev/null 2>&1 || true", | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 174 | let lastState = null; | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 232 | const seedMatches = expected.seed === null || Number(statusData.seed) === expected.seed; | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 238 | if (sawSwitching && statusData.state === "failure") { | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 256 | if (configMatches && statusData.state === "ready" && | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 260 | if (sawSwitching && statusData.state === "failure") { | CODE |
| HIGH⚡ | …ualization/visualization/image_web_viewer_node/page.py | 324 | if (opt && typeof opt === "object") { | CODE |
| HIGH⚡ | …ualization/visualization/image_web_viewer_node/page.py | 326 | o.textContent = String(opt.label || opt.value); | CODE |
| HIGH⚡ | …ualization/visualization/image_web_viewer_node/page.py | 333 | if (current && Array.from(select.options).some(o => o.value === String(current))) { | CODE |
| HIGH | …ualization/visualization/image_web_viewer_node/page.py | 350 | if (stats && stats.episodes > 0) { | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 522 | # Step 1: Crop first to avoid exceeding target dimensions | COMMENT |
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 525 | # Step 2: Calculate padding needed for time and frequency dimensions | COMMENT |
| LOW⚡ | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 529 | # Step 3: Apply padding if needed | COMMENT |
| LOW | …v/models/video_encoders/hf/ltx2/audio_vae/audio_vae.py | 541 | # Step 4: Final safety crop to ensure exact target shape | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 53 | # Step 1: RMSNorm | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 68 | # Step 2: Load scale and shift based on format | COMMENT |
| LOW | lightx2v/models/networks/ltx2/infer/triton_ops.py | 87 | # Step 3: Apply modulation: x_hat * (1 + scale) + shift | COMMENT |
| LOW | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 233 | # Step 1: Cache text KV (called once at generation start) | COMMENT |
| LOW⚡ | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 259 | # Step 2: For non-first chunks, cache context frame KV | COMMENT |
| LOW⚡ | …ghtx2v/models/runners/worldplay/worldplay_ar_runner.py | 264 | # Step 3: Run denoising for this chunk | COMMENT |
| LOW | scripts/server/post_multi_servers.py | 50 | # Step 2: Monitor task status until completion | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 52 | # Step 1: Send to Decoder first (sets up Phase2 receiver) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 59 | # Step 2: Send to Transformer (Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 66 | # Step 3: Send to Encoder (text encoding + Phase1 send) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_t2i.py | 72 | # Step 4: Poll Decoder for completion (image saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 60 | # Step 1: Send to Decoder first (sets up Phase2 receiver) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 67 | # Step 2: Send to Transformer (Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 74 | # Step 3: Send to Encoder (text + image encoding + Phase1 send) | COMMENT |
| LOW⚡ | scripts/server/disagg/qwen/post_qwen_i2i.py | 80 | # Step 4: Poll Decoder for completion (image saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 46 | # Step 1: Send to Decoder first (sets up Phase2 receiver, starts blocking) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 53 | # Step 2: Send to Transformer (sets up Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 60 | # Step 3: Send to Encoder (triggers T5 encoding + Phase1 Mooncake send) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_t2v.py | 66 | # Step 4: Poll Decoder for final completion (video is saved on Decoder node) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 57 | # Step 1: Send to Decoder first (sets up Phase2 receiver, starts blocking) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 64 | # Step 2: Send to Transformer (sets up Phase1 receiver + Phase2 sender) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 71 | # Step 3: Send to Encoder (triggers T5/CLIP/VAE encoding + Phase1 Mooncake send) | COMMENT |
| LOW⚡ | scripts/server/disagg/wan/post_wan_i2v.py | 77 | # Step 4: Poll Decoder for final completion (video is saved on Decoder node) | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/preprocess/pose2d.py | 348 | Load images from various input types. Args: inputs (Union[str, np.ndarray, List[np.ndarray | STRING |
| HIGH | lightx2v/utils/input_info.py | 475 | Calculate num_frames from video duration using the formula: num_frames = (fps * seconds + 3) // 4 * 4 + 1 This | STRING |
| HIGH | …coders/hf/hunyuan3d/autoencoders/surface_extractors.py | 135 | Extract surface mesh using Differentiable Marching Cubes (DMC) algorithm. Args: grid_logit | STRING |
| HIGH | lightx2v/models/networks/worldmirror/utils/geometry.py | 124 | Create a sliding window view of the input array along a specified axis. This function creates a memory-efficie | STRING |
| HIGH | …htx2v/models/networks/worldmirror/utils/visual_util.py | 229 | Converts model predictions to a 3D scene represented as a GLB file. Args: predictions (dict): Dictiona | STRING |
| HIGH | …tx2v/models/networks/worldmirror/models/layers/rope.py | 232 | Applies 2D rotary position embeddings to input tokens. Args: tokens: Input tensor of shape (batch_s | STRING |
| HIGH | …odels/networks/wan/infer/matrix_game2/posemb_layers.py | 71 | Reshape frequency tensor for broadcasting it with another tensor. This function reshapes the frequency tensor | STRING |
| HIGH | …ghtx2v/models/input_encoders/hf/vace/vace_processor.py | 28 | Resize, center crop and normalize for decord loaded video (torch.Tensor type) Parameters: vi | STRING |
| HIGH | lightx2v_kernel/python/lightx2v_kernel/utils.py | 39 | Compute SNR between y_pred(tensor) and y_real(tensor) SNR can be calcualted as following equation: SN | STRING |
| HIGH | …ain/lightx2v_train/model_zoo/native/wan/utils/utils.py | 69 | Convert a string to a boolean. Supported true values: 'yes', 'true', 't', 'y', '1' Supported false values: | STRING |
| HIGH | …tx2v_train/model_zoo/native/wan/utils/qwen_vl_utils.py | 129 | calculate the number of frames for video used for model inputs. Args: ele (dict): a dict contains the confi | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | lightx2v/utils/generate_task_id.py | 22 | # Create an independent random instance | COMMENT |
| MEDIUM | …models/video_encoders/hf/ltx2/video_vae/convolution.py | 124 | # Define the size of the channels after the first convolution | COMMENT |
| MEDIUM | …tx2v/models/networks/worldmirror/models/layers/rope.py | 3 | # This module provides a clean implementation of 2D Rotary Position Embeddings, | COMMENT |
| MEDIUM | …models/networks/worldmirror/models/heads/dense_head.py | 142 | # This module implements the DPT (Dense Prediction Transformer) head as proposed in | STRING |
| MEDIUM | lightx2v/models/networks/neopp/infer/pre_infer.py | 34 | # Create the batch index for each patch (B x patch count) | COMMENT |
| MEDIUM | …x2v/models/networks/hunyuan3d/utils/torchvision_fix.py | 20 | # Create a mock functional_tensor module with the required functions | COMMENT |
| MEDIUM | …x2v/models/networks/hunyuan3d/utils/torchvision_fix.py | 82 | # Create the mock module instance | COMMENT |
| MEDIUM | …x2v/models/input_encoders/hf/hunyuan15/qwen25/model.py | 324 | # Create a deep copy to avoid modifying the original template | COMMENT |
| MEDIUM⚡ | …htx2v_train/lightx2v_train/schedulers/flow_matching.py | 108 | # The following methods are for inference only | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | tools/extract/lora_extractor.py | 213 | # Check if tensor shapes match | COMMENT |
| LOW⚡ | tools/extract/lora_extractor.py | 217 | # Check if tensor data types match | COMMENT |
| LOW⚡ | tools/extract/lora_extractor.py | 224 | # Check if there are actual changes | COMMENT |
| LOW | tools/extract/lora_extractor.py | 250 | # Check if rank is reasonable | COMMENT |
| LOW | tools/convert/converter.py | 387 | # Check if key matches target modules | COMMENT |
| LOW | lightx2v/utils/lora_loader.py | 157 | # Check if down weight exists | COMMENT |
| LOW | …v/models/video_encoders/hf/ltx2/video_vae/video_vae.py | 798 | # Check if current temporal slice overlaps with previous temporal slice | STRING |
| LOW | …2v/models/networks/ltx2/weights/transformer_weights.py | 183 | # Check if tensor parallel is enabled | COMMENT |
| LOW | …odels/networks/worldplay/infer/ar_transformer_infer.py | 751 | # Check if ProPE is enabled | COMMENT |
| LOW | lightx2v/models/networks/worldplay/infer/pre_infer.py | 63 | # Check if we have per-frame timestep from BI runner | COMMENT |
| LOW | lightx2v/models/networks/longcat_image/model.py | 122 | # Check if CFG parallel should be used | COMMENT |
| LOW | …v/models/runners/longcat_image/longcat_image_runner.py | 235 | # Check if prompt rewrite is enabled | COMMENT |
| LOW | lightx2v/common/ops/attn/svg2_attn_utils.py | 940 | # Check if this block needs computation | COMMENT |
| LOW | lightx2v/common/ops/attn/svg2_attn.py | 186 | # Check if block_col_sz and block_row_sz are the same for each head | COMMENT |
| LOW | lightx2v/common/ops/attn/kernels/sla_kernel.py | 240 | # Increment pointers | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | lightx2v_ros/src/visualization/package.xml | 7 | <maintainer email="user@example.com">user</maintainer> | CODE |
| LOW | lightx2v_ros/src/visualization/setup.py | 16 | maintainer_email="user@example.com", | CODE |
| LOW | lightx2v_ros/src/common/package.xml | 7 | <maintainer email="user@example.com">user</maintainer> | CODE |
| LOW | lightx2v_ros/src/common/setup.py | 16 | maintainer_email="user@example.com", | CODE |
| LOW | lightx2v_ros/src/simulator/package.xml | 7 | <maintainer email="user@example.com">user</maintainer> | CODE |
| LOW | lightx2v_ros/src/simulator/setup.py | 16 | maintainer_email="user@example.com", | CODE |
| LOW | lightx2v_ros/src/inference/package.xml | 7 | <maintainer email="user@example.com">user</maintainer> | CODE |
| LOW | lightx2v_ros/src/inference/setup.py | 16 | maintainer_email="user@example.com", | CODE |
| LOW | scripts/platforms/mlu/openai_image/post_i2i_openai.sh | 9 | --api_key "dummy-key" \ | CODE |
| LOW | scripts/platforms/mlu/openai_image/post_t2i_openai.sh | 9 | --api_key "dummy-key" \ | CODE |
| LOW | scripts/server/openai_image/post_i2i_openai.sh | 9 | --api_key "dummy-key" \ | CODE |
| LOW | …ripts/server/openai_image/test_openai_images_client.py | 100 | parser.add_argument("--api_key", type=str, default="dummy-key", help="OpenAI API key placeholder") | CODE |
| LOW | scripts/server/openai_image/post_t2i_openai.sh | 9 | --api_key "dummy-key" \ | CODE |
| LOW | scripts/hidream_o1_image/post_i2i_openai.sh | 11 | --api_key "dummy-key" \ | CODE |
| LOW | scripts/hidream_o1_image/test_openai_images_client.py | 324 | parser.add_argument("--api_key", type=str, default="dummy-key", help="OpenAI API key placeholder") | CODE |
| LOW | scripts/hidream_o1_image/post_t2i_openai.sh | 12 | --api_key "dummy-key" \ | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/preprocess/human_visualization.py | 1263 | CODE | |
| MEDIUM | …tx2v/models/input_encoders/hf/hunyuan3d/conditioner.py | 269 | CODE | |
| MEDIUM | …tx2v/models/input_encoders/hf/hunyuan3d/conditioner.py | 272 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | lightx2v/models/runners/wan/wan_dancer_runner.py | 55 | # two-stage orchestration in run_pipeline and never calls the binding. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …orms/intel_xpu/run_minimax_h3_fl2av_turbo_sla_4step.sh | 13 | # Usage: | COMMENT |
| LOW | …pts/server/disagg/qwen/start_qwen_t2i_decentralized.sh | 9 | # Usage: | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …htx2v/models/runners/worldmirror/worldmirror_runner.py | 324 | # fp8-pertensor auto-quant path consumes. Keys just pass through to | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | lightx2v/models/networks/base_model.py | 686 | This is a basic implementation. Subclasses may override for more | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | lightx2v_ros/src/simulator/simulator/sim/evaluator.py | 68 | def run_task(cfg, task, policy): | CODE |