ace-step/ACE-Step-1.5

36.2

Adjusted Score

36.2

Raw Score

100%

Time Factor

2026-08-29

Last Push

12.4K

Stars

Python

Language

175.3K

Lines of Code

1.2K

Files

4.1K

Pattern Hits

2026-08-29

Scan Date

0.21

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

Longitudinal tracking requires multiple scan runs. Once this repository is re-scanned after new commits land, this chart will visualise how the synthetic code signal evolves over time — enabling you to detect whether AI authorship is growing, stabilising, or being actively corrected by human engineers.

No multi-scan history yet — run the scanner again to build trend data.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 0HIGH 249MEDIUM 538LOW 3312

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 4099 distinct pattern matches across 20 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. 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 diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Hyper-Verbose Identifiers1597 hits · 1759 pts

Severity	File	Line	Snippet	Context
LOW⚡	profile_inference.py	265	def print_time_costs_breakdown(	CODE
LOW	profile_inference.py	1395	def _print_batch_boundary_summary(results: List[Dict]):	CODE
LOW⚡	cli.py	346	def _install_prompt_edit_hook(	CODE
LOW	cli.py	93	def _get_default_checkpoint_dir() -> str:	CODE
LOW	cli.py	224	def _edit_formatted_prompt_via_file(formatted_prompt: str, instruction_path: str) -> str:	CODE
LOW	cli.py	246	def _extract_caption_lyrics_from_formatted_prompt(formatted_prompt: str) -> Tuple[Optional[str], Optional[str]]:	CODE
LOW	cli.py	267	def _extract_instruction_from_formatted_prompt(formatted_prompt: str) -> Optional[str]:	CODE
LOW	cli.py	276	def _extract_cot_metadata_from_formatted_prompt(formatted_prompt: str) -> dict:	CODE
LOW	cli.py	497	def _default_instruction_for_task(task_type: str, tracks: Optional[List[str]] = None) -> str:	CODE
LOW	openrouter/stress_test.py	255	def _get_instrumental_payload(self) -> dict:	CODE
LOW	openrouter/openrouter_api_server.py	320	def _extract_prompt_and_lyrics(messages: List[ChatMessage]) -> tuple[str, str, str, List[str]]:	CODE
LOW	openrouter/client_test.py	131	def test_natural_language_mode(base_url: str, api_key: Optional[str] = None) -> bool:	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	89	def test_includes_pytorch_cache_when_device_free_is_zero(self):	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	109	def test_sums_device_free_and_allocator_cache(self):	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	123	def test_returns_zero_when_fully_allocated_no_cache(self):	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	136	def test_returns_zero_when_cuda_unavailable(self):	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	146	def test_debug_cap_clamps_effective_free(self):	CODE
LOW⚡	acestep/launcher_compat_test.py	34	def test_legacy_arch_missing_requires_fix(self) -> None:	CODE
LOW⚡	acestep/launcher_compat_test.py	42	def test_legacy_arch_present_skips_fix(self) -> None:	CODE
LOW⚡	acestep/launcher_compat_test.py	50	def test_modern_gpu_skips_fix_even_if_arch_not_listed(self) -> None:	CODE
LOW	acestep/launcher_compat_test.py	66	def test_determine_returns_probe_failed_on_exception(self) -> None:	CODE
LOW	acestep/launcher_compat_test.py	77	def test_probe_exit_code_maps_to_decision(self) -> None:	CODE
LOW	acestep/llm_backend_compat.py	7	def _has_working_triton_installation() -> bool:	CODE
LOW	acestep/llm_backend_compat.py	17	def get_vllm_preflight_warning(*, device: str, platform: str \| None = None) -> str \| None:	CODE
LOW	acestep/audio_utils_uuid_test.py	32	def test_different_lora_states_produce_different_uuids(self):	CODE
LOW	acestep/audio_utils_uuid_test.py	62	def test_different_lora_scale_produces_different_uuids(self):	CODE
LOW	acestep/audio_utils_uuid_test.py	86	def test_same_params_produce_same_uuid(self):	CODE
LOW	acestep/audio_utils_uuid_test.py	102	def test_uuid_format_is_valid(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	126	def test_lora_weights_hash_empty_when_no_lora(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	131	def test_lora_weights_hash_empty_when_use_lora_false(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	136	def test_different_lora_files_produce_different_hashes(self):	CODE
LOW	acestep/audio_utils_uuid_test.py	172	def test_same_lora_file_produces_same_hash(self):	CODE
LOW	acestep/audio_utils_uuid_test.py	192	def test_lora_weights_hash_differentiates_uuids(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	224	def test_user_codes_preserved_when_lm_returns_empty(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	232	def test_lm_codes_overwrite_when_non_empty(self):	CODE
LOW⚡	acestep/audio_utils_uuid_test.py	239	def test_user_codes_preserved_when_lm_list_empty(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	21	def test_missing_sidecar_returns_none(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	26	def test_loads_relative_latent_file_from_sidecar(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	43	def test_loads_source_seed_from_sidecar(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	63	def test_resample_matching_source_seed_only_changes_collisions(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	71	def test_loads_sidecar_from_gradio_outputs_by_audio_basename(self):	CODE
LOW	acestep/inference_repaint_cache_test.py	92	def test_gradio_output_sidecar_search_uses_literal_latest_match(self):	CODE
LOW	acestep/openrouter_adapter.py	321	def _to_generate_music_request(	CODE
LOW	acestep/openrouter_adapter.py	416	def _build_openrouter_response(	CODE
LOW	acestep/openrouter_adapter.py	473	async def _openrouter_stream_generator(	CODE
LOW	acestep/openrouter_adapter.py	587	def _get_model_name_from_path(config_path: str) -> str:	CODE
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	228	def test_codes_generation_only_samples_valid_tokens(self):	CODE
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	238	def test_non_codes_state_samples_from_full_vocab(self):	CODE
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	248	def test_codes_generation_cfg_applied_to_valid_indices_only(self):	CODE
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	396	def test_generation_stops_when_all_sequences_finish(self):	CODE
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	431	def test_finished_sequences_are_forced_to_eos(self):	CODE
LOW	acestep/llm_inference_cfg_fixes_test.py	44	def _make_constrained_processor(state: "FSMState", vocab_size: int = 10) -> MagicMock:	CODE
LOW	acestep/llm_inference_cfg_fixes_test.py	324	def test_cfg_nan_replaced_with_neg_inf(self):	CODE
LOW⚡	acestep/gpu_config.py	124	def get_cuda_device_capability(device_index: int = 0) -> Optional[Tuple[int, int]]:	CODE
LOW	acestep/gpu_config.py	273	def _apply_lm_backend_compatibility_overrides(config: GPUConfig) -> GPUConfig:	CODE
LOW	acestep/gpu_config.py	940	def find_best_lm_model_on_disk(	CODE
LOW	acestep/gpu_config.py	1188	def get_effective_free_vram_gb(device_index: int = 0) -> float:	CODE
LOW	acestep/direct_conditioning_tasks_test.py	139	def test_dit_receives_source_audio_and_params_conditioning(self):	CODE
LOW	acestep/direct_conditioning_tasks_test.py	147	def test_no_audio_codes_pre_empt_source_audio(self):	CODE
LOW⚡	acestep/direct_conditioning_tasks_test.py	161	def test_dit_receives_source_audio_and_params_conditioning(self):	CODE
1537 more matches not shown…

Decorative Section Separators499 hits · 1566 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	docker-compose.jetson.yml	1	# =============================================================================	COMMENT
MEDIUM⚡	docker-compose.jetson.yml	3	# =============================================================================	COMMENT
MEDIUM	docker-compose.jetson.yml	31	# =============================================================================	COMMENT
MEDIUM⚡	profile_inference.py	260	# =============================================================================	COMMENT
MEDIUM⚡	profile_inference.py	262	# =============================================================================	COMMENT
MEDIUM⚡	profile_inference.py	1719	# =============================================================================	COMMENT
MEDIUM⚡	profile_inference.py	1721	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	85	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	87	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	140	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	142	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	194	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	196	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	420	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	422	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	538	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	540	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	716	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	718	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1521	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1523	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1572	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1574	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1623	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1625	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1683	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1685	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1794	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	1796	# =============================================================================	COMMENT
MEDIUM	profile_inference.py	2095	# =============================================================================	STRING
MEDIUM	profile_inference.py	2097	# =============================================================================	STRING
MEDIUM	requirements-rocm.txt	2	# =================================================	COMMENT
MEDIUM	train.py	31	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	train.py	33	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	train.py	144	# ===========================================================================	COMMENT
MEDIUM⚡	train.py	146	# ===========================================================================	COMMENT
MEDIUM	train.py	265	# ===========================================================================	COMMENT
MEDIUM	train.py	267	# ===========================================================================	COMMENT
MEDIUM⚡	docker-compose.yml	1	# =============================================================================	COMMENT
MEDIUM⚡	docker-compose.yml	3	# =============================================================================	COMMENT
MEDIUM⚡	docker-compose.yml	15	# =============================================================================	COMMENT
MEDIUM⚡	requirements-sidestep.txt	1	# ╔══════════════════════════════════════════════════════════════╗	COMMENT
MEDIUM⚡	requirements-sidestep.txt	4	# ╚══════════════════════════════════════════════════════════════╝	COMMENT
MEDIUM⚡	requirements-sidestep.txt	9	# ── Required for Side-Step CLI ────────────────────────────────	COMMENT
MEDIUM⚡	requirements-sidestep.txt	12	# ── Required for Side-Step TUI ────────────────────────────────	COMMENT
MEDIUM⚡	requirements-sidestep.txt	20	# ── Optional: Prodigy adaptive optimizer (auto-tunes LR) ─────	COMMENT
MEDIUM	openrouter/stress_test.py	37	# =============================================================================	COMMENT
MEDIUM	openrouter/stress_test.py	39	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	52	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	54	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	65	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	67	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	96	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	98	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	230	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	232	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	494	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	496	# =============================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	532	# =================================================================	COMMENT
MEDIUM	openrouter/openrouter_api_server.py	534	# =================================================================	COMMENT
439 more matches not shown…

Cross-File Repetition179 hits · 895 pts

Severity	File	Line	Snippet	Context
HIGH	…i/gradio/events/results/audio_playback_updates_test.py	0	load target module directly by file path for isolated testing.	STRING
HIGH	…/ui/gradio/interfaces/audio_player_preferences_test.py	0	load target module directly by file path for isolated testing.	STRING
HIGH	acestep/ui/gradio/interfaces/user_preferences_test.py	0	load target module directly by file path for isolated testing.	STRING
HIGH	…estep/ui/gradio/events/results/generation_info_test.py	0	load target module directly by file path for isolated testing.	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	general 4d attention mask generator compatible with cpu/mac/sdpa and eager mode. supports use cases: 1. causal full: is_	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	pack two sequences by concatenating and sorting them based on mask values. args: hidden1: first hidden states tensor of	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	sample timestep t and r for flow matching training. args: batch_size: batch size device: device to create tensors on dty	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	timestep embedding module for diffusion models. converts timestep values into high-dimensional embeddings using sinusoid	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	create sinusoidal timestep embeddings. args: t: a 1-d tensor of n indices, one per batch element. these may be fractiona	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	multi-headed attention module for acestep model. implements the attention mechanism from 'attention is all you need' pap	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	encoder layer for acestep model. consists of self-attention and mlp (feed-forward) sub-layers with residual connections.	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	dit (diffusion transformer) layer for acestep model. implements a transformer layer with three main components: 1. self-	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	acestep/models/sft/modeling_acestep_v15_base.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	acestep/models/base/modeling_acestep_v15_base.py	0	initialize weights for different module types. todo: support separate initialization for encoders and decoders.	STRING
HIGH	acestep/models/turbo/modeling_acestep_v15_turbo.py	0	encoder for processing lyric text embeddings. encodes lyric text hidden states using a transformer encoder architecture	STRING
HIGH	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	0	encoder for processing lyric text embeddings. encodes lyric text hidden states using a transformer encoder architecture	STRING
119 more matches not shown…

Unused Imports687 hits · 612 pts

Severity	File	Line	Context
LOW	profile_inference.py	74	CODE
LOW	profile_inference.py	74	CODE
LOW	profile_inference.py	107	CODE
LOW	profile_inference.py	860	CODE
LOW	profile_inference.py	1735	CODE
LOW	cli.py	1063	CODE
LOW	train.py	25	CODE
LOW	openrouter/stress_test.py	32	CODE
LOW	openrouter/openrouter_api_server.py	14	CODE
LOW	openrouter/openrouter_api_server.py	20	CODE
LOW	openrouter/openrouter_api_server.py	25	CODE
LOW	openrouter/openrouter_api_server.py	38	CODE
LOW	openrouter/openrouter_api_server.py	39	CODE
LOW	openrouter/client_test.py	17	CODE
LOW	acestep/gpu_config_effective_free_vram_test.py	14	CODE
LOW	acestep/launcher_compat_test.py	3	CODE
LOW	acestep/cli_args.py	3	CODE
LOW	acestep/openrouter_adapter.py	14	CODE
LOW	acestep/llm_inference_cfg_fixes_test.py	13	CODE
LOW	acestep/openrouter_models.py	7	CODE
LOW	acestep/llm_backend_compat_test.py	3	CODE
LOW	acestep/llm_inference.py	28	CODE
LOW	acestep/llm_inference.py	2928	CODE
LOW	acestep/llm_inference.py	666	CODE
LOW	acestep/llm_inference.py	2965	CODE
LOW	acestep/llm_inference.py	2965	CODE
LOW	acestep/inference_seed_fallback_test.py	9	CODE
LOW	acestep/gpu_config_lm_memory_ratio_test.py	28	CODE
LOW	acestep/acestep_v15_pipeline_test.py	3	CODE
LOW	acestep/audio_utils_test.py	3	CODE
LOW	acestep/audio_utils_test.py	7	CODE
LOW	acestep/audio_utils.py	11	CODE
LOW	acestep/audio_utils.py	18	CODE
LOW	acestep/api_server.py	16	CODE
LOW	acestep/acestep_v15_pipeline.py	52	CODE
LOW	acestep/acestep_v15_pipeline.py	52	CODE
LOW	acestep/acestep_v15_pipeline.py	52	CODE
LOW	acestep/acestep_v15_pipeline.py	74	CODE
LOW	acestep/acestep_v15_pipeline.py	74	CODE
LOW	acestep/acestep_v15_pipeline.py	74	CODE
LOW	acestep/cli_args_test.py	3	CODE
LOW	acestep/inference.py	15	CODE
LOW	acestep/inference.py	24	CODE
LOW	acestep/inference.py	24	CODE
LOW	acestep/inference.py	24	CODE
LOW	acestep/inference.py	24	CODE
LOW	acestep/inference.py	24	CODE
LOW	acestep/model_downloader_test.py	5	CODE
LOW	acestep/launcher_compat.py	3	CODE
LOW	acestep/constrained_logits_processor.py	24	CODE
LOW	acestep/constrained_logits_processor.py	24	CODE
LOW	acestep/constrained_logits_processor.py	30	CODE
LOW	acestep/constrained_logits_processor.py	30	CODE
LOW	acestep/launcher_legacy_torch_fix_test.py	3	CODE
LOW	acestep/dataset_handler.py	11	CODE
LOW	acestep/dataset_handler.py	11	CODE
LOW	acestep/dataset_handler.py	11	CODE
LOW	acestep/text2music_src_audio_test.py	14	CODE
LOW	acestep/text2music_src_audio_test.py	15	CODE
LOW	acestep/text2music_src_audio_test.py	16	CODE
627 more matches not shown…

Excessive Try-Catch Wrapping421 hits · 449 pts

Severity	File	Line	Snippet	Context
LOW⚡	profile_inference.py	255	except Exception as e:	CODE
LOW⚡	profile_inference.py	1715	except Exception:	CODE
LOW	profile_inference.py	920	except Exception as e:	CODE
LOW	profile_inference.py	1005	except Exception as e:	CODE
LOW	profile_inference.py	1217	except Exception:	CODE
LOW	profile_inference.py	1225	except Exception:	CODE
MEDIUM	cli.py	991	print(f"Error saving configuration: {e}. Please try again.")	CODE
LOW⚡	cli.py	335	except Exception:	CODE
LOW⚡	cli.py	342	except Exception:	CODE
LOW	cli.py	37	except Exception:	CODE
LOW	cli.py	229	except Exception as e:	CODE
LOW	cli.py	241	except Exception as e:	CODE
LOW	cli.py	316	except Exception:	CODE
LOW	cli.py	477	except Exception:	CODE
LOW	cli.py	990	except Exception as e:	CODE
LOW	cli.py	1152	except Exception as e:	CODE
LOW	cli.py	1337	except Exception as e:	CODE
LOW	cli.py	1648	except Exception as e:	CODE
LOW⚡	train.py	134	except Exception as exc:	CODE
LOW	train.py	188	except Exception as exc:	CODE
LOW	train.py	234	except Exception as exc:	CODE
LOW	generate_examples.py	130	except Exception as e:	CODE
LOW	openrouter/stress_test.py	233	except Exception as e:	CODE
LOW	openrouter/stress_test.py	305	except Exception as e:	CODE
LOW	openrouter/stress_test.py	726	except Exception as e:	STRING
LOW⚡	openrouter/openrouter_api_server.py	967	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	364	except Exception:	CODE
LOW	openrouter/openrouter_api_server.py	585	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	743	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	785	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	825	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	1019	except Exception as e:	CODE
LOW	openrouter/openrouter_api_server.py	1082	except Exception as e:	CODE
LOW⚡	openrouter/client_test.py	494	except Exception as e:	CODE
LOW	openrouter/client_test.py	106	except Exception as e:	CODE
LOW	openrouter/client_test.py	126	except Exception as e:	CODE
LOW	openrouter/client_test.py	183	except Exception as e:	CODE
LOW	openrouter/client_test.py	243	except Exception as e:	CODE
LOW	openrouter/client_test.py	284	except Exception as e:	CODE
LOW	openrouter/client_test.py	319	except Exception as e:	CODE
LOW	openrouter/client_test.py	411	except Exception as e:	CODE
LOW	openrouter/client_test.py	454	except Exception as e:	CODE
LOW	openrouter/client_test.py	479	except Exception as e:	CODE
LOW	acestep/openrouter_adapter.py	165	except Exception:	CODE
LOW	acestep/openrouter_adapter.py	175	except Exception:	CODE
LOW	acestep/openrouter_adapter.py	674	except Exception as e:	CODE
LOW	acestep/openrouter_adapter.py	770	except Exception as exc:	CODE
LOW⚡	acestep/gpu_config.py	57	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	67	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	77	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	87	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	120	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	131	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	1253	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	1261	except Exception:	CODE
LOW⚡	acestep/gpu_config.py	1265	except Exception:	CODE
LOW	acestep/gpu_config.py	107	except Exception:	CODE
LOW	acestep/gpu_config.py	530	except Exception as e:	CODE
LOW	acestep/gpu_config.py	606	except Exception as e:	CODE
LOW	acestep/gpu_config.py	623	except Exception:	CODE
361 more matches not shown…

Deep Nesting225 hits · 190 pts

Severity	File	Line	Context
LOW	profile_inference.py	116	CODE
LOW	profile_inference.py	543	CODE
LOW	profile_inference.py	1022	CODE
LOW	profile_inference.py	2100	CODE
LOW	cli.py	657	CODE
LOW	cli.py	1000	CODE
LOW	generate_examples.py	20	CODE
LOW	openrouter/stress_test.py	523	CODE
LOW	openrouter/stress_test.py	159	CODE
LOW	openrouter/openrouter_api_server.py	320	CODE
LOW	openrouter/openrouter_api_server.py	498	CODE
LOW	openrouter/openrouter_api_server.py	631	CODE
LOW	openrouter/openrouter_api_server.py	704	CODE
LOW	openrouter/openrouter_api_server.py	790	CODE
LOW	openrouter/client_test.py	131	CODE
LOW	openrouter/client_test.py	190	CODE
LOW	openrouter/client_test.py	326	CODE
LOW	acestep/openrouter_adapter.py	236	CODE
LOW	acestep/openrouter_adapter.py	473	CODE
LOW	acestep/openrouter_adapter.py	575	CODE
LOW	acestep/openrouter_adapter.py	654	CODE
LOW	acestep/llm_inference_cfg_fixes_test.py	78	CODE
LOW	acestep/llm_inference_cfg_fixes_test.py	122	CODE
LOW	acestep/gpu_config.py	476	CODE
LOW	acestep/gpu_config.py	743	CODE
LOW	acestep/gpu_config.py	976	CODE
LOW	acestep/gpu_config.py	1077	CODE
LOW	acestep/gpu_config.py	1188	CODE
LOW	acestep/gpu_config.py	1470	CODE
LOW	acestep/llm_inference.py	91	CODE
LOW	acestep/llm_inference.py	125	CODE
LOW	acestep/llm_inference.py	499	CODE
LOW	acestep/llm_inference.py	874	CODE
LOW	acestep/llm_inference.py	985	CODE
LOW	acestep/llm_inference.py	1145	CODE
LOW	acestep/llm_inference.py	1282	CODE
LOW	acestep/llm_inference.py	2216	CODE
LOW	acestep/llm_inference.py	2362	CODE
LOW	acestep/llm_inference.py	2618	CODE
LOW	acestep/llm_inference.py	2792	CODE
LOW	acestep/llm_inference.py	3036	CODE
LOW	acestep/llm_inference.py	3397	CODE
LOW	acestep/llm_inference.py	4099	CODE
LOW	acestep/llm_inference.py	4157	CODE
LOW	acestep/llm_inference.py	2852	CODE
LOW	acestep/model_downloader.py	442	CODE
LOW	acestep/model_downloader.py	864	CODE
LOW	acestep/audio_utils.py	451	CODE
LOW	acestep/audio_utils.py	511	CODE
LOW	acestep/audio_utils.py	197	CODE
LOW	acestep/acestep_v15_pipeline.py	144	CODE
LOW	acestep/inference.py	514	CODE
LOW	acestep/constrained_logits_processor.py	548	CODE
LOW	acestep/constrained_logits_processor.py	676	CODE
LOW	acestep/constrained_logits_processor.py	788	CODE
LOW	acestep/constrained_logits_processor.py	851	CODE
LOW	acestep/constrained_logits_processor.py	913	CODE
LOW	acestep/constrained_logits_processor.py	953	CODE
LOW	acestep/constrained_logits_processor.py	1079	CODE
LOW	acestep/constrained_logits_processor.py	1335	CODE
165 more matches not shown…

Docstring Block Structure31 hits · 155 pts

Severity	File	Line	Snippet	Context
HIGH	acestep/cli_args.py	16	Parse ``--quantization`` values from CLI input. Args: value: Raw CLI value. Returns: Canonical	STRING
HIGH	acestep/model_downloader.py	711	Resolve a VAE variant id (or absolute path) to its on-disk directory. Args: checkpoint_dir: Root checkpoint	STRING
HIGH	…io/events/wiring/generation_batch_navigation_wiring.py	176	Register previous/next batch navigation and background pre-generation wiring. Args: context (GenerationWiri	STRING
HIGH	…estep/ui/gradio/events/wiring/generation_run_wiring.py	30	Proxy passthrough to `res_h.generate_with_batch_management`. Args: *args (Any): Positional passthro	STRING
HIGH	…/third_parts/nano-vllm/nanovllm/engine/model_runner.py	30	Encode IPC arguments to UTF-8 JSON bytes for shared-memory transport. Replaces ``pickle`` to avoid arbitrary code e	STRING
HIGH	…/third_parts/nano-vllm/nanovllm/engine/model_runner.py	84	Find an available port starting from start_port. Args: start_port: The starting port number to check	STRING
HIGH	acestep/core/generation/handler/lyric_timestamp.py	28	Generate LRC timestamps by aligning decoder cross-attention to lyric tokens. Args: pred_latent (tor	STRING
HIGH	…estep/core/generation/handler/generate_music_decode.py	26	Collect decode inputs and validate raw diffusion latents. Args: outputs: ``service_generate`` outpu	STRING
HIGH	…e/generation/handler/init_service_loader_components.py	27	Load the VAE checkpoint and return its resolved path. Args: checkpoint_dir: Root checkpoint directo	STRING
HIGH	…e/generation/handler/init_service_loader_components.py	69	Load the text tokenizer and embedding model, then return its path. Args: checkpoint_dir: Root check	STRING
HIGH	acestep/core/generation/handler/mlx_vae_init_test.py	10	Load a handler mixin module for isolated tests. Args: filename: Module filename in the current test directo	STRING
HIGH	acestep/core/generation/handler/mlx_vae_init_test.py	88	Build fake ``mlx.core`` and compile-call tracking for tests. Args: raise_compile: Whether inner ``_compile`	STRING
HIGH	acestep/core/generation/handler/lyric_score.py	27	Calculate lyric alignment scores for pure-noise and regressed-latent inputs. Args: pred_latent (tor	STRING
HIGH	acestep/core/generation/handler/io_audio.py	14	Read an audio file, with torchaudio fallback for formats unsupported by soundfile. soundfile (libsndfile) supports	STRING
HIGH	acestep/core/generation/handler/generate_music.py	252	Generate audio from text/reference inputs and return response payload. Args: captions: Text prompt	STRING
HIGH	acestep/training/data_module.py	544	Load a dataset from JSON file. Args: json_path: Path to the JSON dataset file. Returns: Tuple	STRING
HIGH	acestep/training/path_safety.py	54	Validate and normalise a user-provided path. The returned path is guaranteed to live under base (or the globa	STRING
HIGH	acestep/training/path_safety.py	97	Open a file after validating its path. Convenience wrapper around ``safe_path`` + ``open``. Args: user	STRING
HIGH	acestep/api/job_analysis_runtime.py	19	Run analysis-only branches and return response payload when handled. Args: req: Generation request object.	STRING
HIGH	acestep/api/job_generation_runtime.py	22	Run music generation and aggregate outputs across sequential MPS cover slices. Args: req: Generation reques	STRING
HIGH	acestep/api/llm_generation_inputs.py	44	Resolve LLM flags and prepare caption/lyrics/metadata inputs for generation. Args: app_state: FastAPI app s	STRING
HIGH	acestep/api/http/auth.py	26	Validate request auth from body ``ai_token`` or Authorization header. Args: body: Parsed request payload di	STRING
HIGH	acestep/api/http/model_init_service.py	66	Initialize DiT and optional LM models. Args: app_state: FastAPI application state carrying handler objects	STRING
HIGH	acestep/api/http/release_task_request_parser.py	49	Parse ``/release_task`` request body into request model and temp-file list. Args: request: FastAPI request	STRING
HIGH	acestep/api/http/release_task_audio_paths.py	15	Validate user-supplied audio path and block unsafe filesystem traversal. Args: path: User-supplied path val	STRING
HIGH	acestep/api/http/release_task_audio_paths.py	50	Persist uploaded audio file to a temporary location. Args: upload: Uploaded file wrapper from Starlette/Fas	STRING
HIGH	acestep/training_v2/ui/presets.py	250	Save a preset to the project-local presets directory. Presets are saved to ``./presets/`` (relative to the current	STRING
HIGH	acestep/training_v2/ui/prompt_helpers.py	80	Display a numbered menu and return the chosen key. Args: title: Prompt text. options: List of ``(ke	STRING
HIGH	acestep/training_v2/ui/prompt_helpers.py	157	Ask for a single value with an optional default. Args: label: Prompt text. default: Default value (	STRING
HIGH	acestep/training_v2/ui/flows_train.py	143	Interactive wizard for training (fixed or vanilla). Args: mode: Training mode ('fixed' or 'vanilla').	STRING
HIGH	acestep/training_v2/cli/validation.py	61	Resolve target modules based on attention type selection. Args: target_modules: List of module patterns (e.	STRING

Cross-Language Confusion23 hits · 140 pts

Severity	File	Line	Snippet	Context
HIGH	acestep/llm_inference.py	800	logger.error("nano-vllm is not installed. Please install it using 'cd acestep/third_parts/nano-vllm && pip i	CODE
HIGH	acestep/llm_inference.py	801	return "❌ nano-vllm is not installed. Please install it using 'cd acestep/third_parts/nano-vllm && pip insta	CODE
HIGH	…step/ui/gradio/events/wiring/results_display_wiring.py	30	pathString = current_audio.path \|\| current_audio.name \|\| current_audio.url \|\| "";	CODE
HIGH	…step/ui/gradio/events/wiring/results_display_wiring.py	49	const fPath = f.name \|\| f.path \|\| "";	CODE
HIGH	…step/ui/gradio/events/wiring/results_display_wiring.py	65	a.href = f.url \|\| f.data;	CODE
HIGH	acestep/ui/gradio/interfaces/user_preferences.py	112	skip_sentinel = f"new Array({num_outputs}).fill(null)"	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	130	if (!(k in prefs)) return null;	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	135	if (expected && typeof v !== expected) {{	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	138	return null;	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	146	else return null;	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	151	if (result.every(v => v === null)) return SKIP;	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences.py	159	result.push(mp3, mp3, mp3);	CODE
HIGH	acestep/ui/gradio/interfaces/user_preferences.py	173	- First ``len(PREF_KEYS)`` elements are preference values (or null).	STRING
HIGH	acestep/ui/gradio/interfaces/user_preferences.py	174	- Next 3 elements are mp3 visibility booleans (or null):	STRING
HIGH⚡	…/ui/gradio/interfaces/audio_player_preferences_test.py	62	self.assertIn("if (value === null \|\| value === undefined \|\| value === \"\")", script)	CODE
HIGH⚡	…/ui/gradio/interfaces/audio_player_preferences_test.py	63	self.assertIn("if (preferredVolume === null)", script)	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences_test.py	152	self.assertIn("fill(null)", js)	CODE
HIGH⚡	acestep/ui/gradio/interfaces/user_preferences_test.py	161	self.assertIn('result.push(', js)	CODE
HIGH⚡	acestep/models/common/flow_edit_helpers_test.py	117	out = pack_for_cfg(self.enc_hs, self.enc_am, self.ctx, self.attn, self.null, do_cfg=False)	CODE
HIGH⚡	acestep/models/common/flow_edit_helpers_test.py	124	out = pack_for_cfg(self.enc_hs, self.enc_am, self.ctx, self.attn, self.null, do_cfg=True)	CODE
HIGH	acestep/models/common/flow_edit_helpers.py	50	"""Reduce a ``(cond, null)``-packed prediction to a guided velocity.	STRING
HIGH	acestep/models/common/flow_edit_helpers.py	184	torch.cat([enc_hs, null], dim=0),	CODE
HIGH	acestep/training_v2/timestep_sampling.py	117	# Per-sample mask: 0 = drop condition (replace with null), 1 = keep	COMMENT

Structural Annotation Overuse99 hits · 137 pts

Severity	File	Line	Snippet	Context
LOW	CONTRIBUTING.md	65	### Step 1: Commit-Scoped Review (First Pass)	COMMENT
LOW⚡	CONTRIBUTING.md	84	### Step 2: Validate Findings	COMMENT
LOW⚡	CONTRIBUTING.md	94	### Step 3: Apply Minimal Fixes	COMMENT
LOW⚡	CONTRIBUTING.md	100	### Step 4: PR-Scoped Review (Second Pass)	COMMENT
LOW	CONTRIBUTING.md	114	### Step 5: Write Reviewer Responses	COMMENT
LOW⚡	openrouter/openrouter_api_server.py	963	# Step 1: Run LM sample generation	COMMENT
LOW	openrouter/openrouter_api_server.py	992	# Step 2: Run audio generation with heartbeats	COMMENT
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	438	# Step 0: output EOS; step 1+: output token 5	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	719	# Step 1: Encode full string "keyscale: {keyscale}" (with space, as tokenizer sees it)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	723	# Step 2: Find where context ends in full_token_ids	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	738	# Step 3: Extract keyscale value tokens (everything after context)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	741	# Step 4: Verify we extracted some tokens (sanity check)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	747	# Step 5: Verify first token is a note (A-G)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	759	# Step 6: Build prefix mappings from keyscale value tokens	COMMENT
LOW⚡	acestep/ui/gradio/i18n/en.json	528	"training_dataset": "## Dataset Builder Tutorial\n\n### Step 1: Load or Scan\n- Load: Enter path to existing dat	CODE
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	329	# Step 1: Self-attention with AdaLN	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	339	# Step 2: Cross-attention	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	350	# Step 3: MLP with AdaLN	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
39 more matches not shown…

Redundant / Tautological Comments66 hits · 85 pts

Severity	File	Line	Snippet	Context
LOW	install_uv.sh	27	# Check if uv is already installed	COMMENT
LOW	profile_inference.py	524	# Print results	COMMENT
LOW	start_api_server.sh	121	# Check if uv is installed	COMMENT
LOW	start_api_server.sh	211	# Check if virtual environment exists	COMMENT
LOW	start_gradio_ui_macos_manual.sh	356	# Check if uv is installed	COMMENT
LOW	start_gradio_ui_macos_manual.sh	406	# Check if virtual environment exists	COMMENT
LOW	train.py	245	# Display results	COMMENT
LOW	start_gradio_ui.sh	212	# Check if uv is installed	COMMENT
LOW	start_gradio_ui.sh	328	# Check if virtual environment exists	COMMENT
LOW	start_gradio_ui_macos.sh	302	# Check if uv is installed	COMMENT
LOW	start_gradio_ui_macos.sh	352	# Check if virtual environment exists	COMMENT
LOW	start_gradio_ui_manual.sh	333	# Check if uv is installed	COMMENT
LOW	start_gradio_ui_manual.sh	404	# Check if virtual environment exists	COMMENT
LOW	start_api_server_macos.sh	198	# Check if uv is installed	COMMENT
LOW	start_api_server_macos.sh	243	# Check if virtual environment exists	COMMENT
LOW	check_update.sh	40	# Check if this is a git repository	COMMENT
LOW	check_update.sh	137	# Check if local is behind remote	COMMENT
LOW	check_update.sh	236	# Check if any untracked files conflict with incoming changes	COMMENT
LOW	openrouter/openrouter_api_server.py	644	# Check if model is initialized	COMMENT
LOW	acestep/gpu_config.py	1224	# Check if a per-process memory fraction has been set	COMMENT
LOW	acestep/gpu_config.py	1407	# Check if model size is in available models	COMMENT
LOW	acestep/llm_inference.py	1688	# Check if user provided a meaningful negative prompt (not the default)	COMMENT
LOW	acestep/llm_inference.py	2889	# Check if this is a new field (no leading spaces and contains ':')	COMMENT
LOW	acestep/llm_inference.py	2979	# Check if keys need "model." prefix by inspecting first key	COMMENT
LOW	acestep/llm_inference.py	3287	# Check if all items are done	COMMENT
LOW	acestep/model_downloader.py	646	# Check if this is a known LM model	COMMENT
LOW	acestep/model_downloader.py	648	# Check if it might be a variant name	COMMENT
LOW	acestep/model_downloader.py	689	# Check if this is the default turbo model (part of main)	COMMENT
LOW	acestep/model_downloader.py	693	# Check if it's a known sub-model	COMMENT
LOW	acestep/inference.py	747	# Check if LM generation failed	COMMENT
LOW	acestep/inference.py	918	# Check if generation failed	COMMENT
LOW	acestep/inference.py	1159	# Check if LLM is initialized	COMMENT
LOW	acestep/inference.py	1183	# Check if LLM returned empty metadata (error case)	COMMENT
LOW	acestep/inference.py	1336	# Check if LLM is initialized	COMMENT
LOW	acestep/inference.py	1358	# Check if LLM returned empty metadata (error case)	COMMENT
LOW	acestep/inference.py	1510	# Check if LLM is initialized	COMMENT
LOW	acestep/inference.py	1532	# Check if LLM returned empty metadata (error case)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	751	# Check if first token starts with a note (A-G, case insensitive, with optional leading space)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	1804	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	1260	# Check if new_prefix is a valid prefix in the active trie	COMMENT
LOW	acestep/constrained_logits_processor.py	1389	# Check if decoded token matches the prefix start (allowing for space prefixes)	COMMENT
LOW	acestep/constrained_logits_processor.py	1424	# Check if this digit could lead to a valid final value	COMMENT
LOW	acestep/constrained_logits_processor.py	1433	# Check if we can still reach min_val	COMMENT
LOW	acestep/constrained_logits_processor.py	1597	# Check if input contains </think> token sequence	COMMENT
LOW	acestep/constrained_logits_processor.py	1750	# Check if we have tokens in queue for user-provided field	COMMENT
LOW	acestep/constrained_logits_processor.py	1763	# Check if we should stop at reasoning (after </think> tag)	COMMENT
LOW	acestep/constrained_logits_processor.py	1766	# Check if the next token would complete the fixed string	COMMENT
LOW	acestep/constrained_logits_processor.py	1835	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	1848	# Check if we should transition after a newline (non-indented line = new field)	COMMENT
LOW	acestep/constrained_logits_processor.py	1900	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	1940	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	1983	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	1996	# Check if current token sequence is complete (allows newline)	COMMENT
LOW	acestep/constrained_logits_processor.py	2019	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	2066	# Check if current token sequence is complete (allows newline)	COMMENT
LOW	acestep/constrained_logits_processor.py	2083	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	acestep/constrained_logits_processor.py	2096	# Check if current token sequence is complete (allows newline)	COMMENT
LOW	acestep/constrained_logits_processor.py	2204	# Check if we've completed the fixed string	COMMENT
LOW	acestep/constrained_logits_processor.py	2273	# Check if we've completed a field name (detected colon)	COMMENT
LOW	acestep/ui/gradio/interfaces/__init__.py	62	# Check if running in service mode (hide training tab)	COMMENT
6 more matches not shown…

Magic Placeholder Names16 hits · 78 pts

Severity	File	Line	Snippet	Context
HIGH	openrouter/client_test.py	9	python -m openrouter.test_client --api-key your-api-key	STRING
HIGH	.claude/skills/acestep-docs/api/API.md	43	"ai_token": "your-api-key",	CODE
HIGH	.claude/skills/acestep-docs/api/API.md	53	-H 'Authorization: Bearer your-api-key' \	CODE
HIGH	.claude/skills/acestep-docs/api/Openrouter_API.md	28	Authorization: Bearer <your-api-key>	CODE
HIGH	docs/ja/API.md	45	"ai_token": "your-api-key",	CODE
HIGH	docs/ja/API.md	55	-H 'Authorization: Bearer your-api-key' \	CODE
HIGH	docs/ja/Openrouter_API_DOC.md	29	Authorization: Bearer <your-api-key>	CODE
HIGH	docs/zh/API.md	45	"ai_token": "your-api-key",	CODE
HIGH	docs/zh/API.md	55	-H 'Authorization: Bearer your-api-key' \	CODE
HIGH	docs/zh/Openrouter_API_DOC.md	29	Authorization: Bearer <your-api-key>	CODE
HIGH	docs/ko/API.md	45	"ai_token": "your-api-key",	CODE
HIGH	docs/ko/API.md	55	-H 'Authorization: Bearer your-api-key' \	CODE
HIGH	docs/ko/Openrouter_API_DOC.md	29	Authorization: Bearer <your-api-key>	CODE
HIGH	docs/en/API.md	46	"ai_token": "your-api-key",	CODE
HIGH	docs/en/API.md	56	-H 'Authorization: Bearer your-api-key' \	CODE
HIGH	docs/en/Openrouter_API_DOC.md	29	Authorization: Bearer <your-api-key>	CODE

Verbosity Indicators56 hits · 70 pts

Severity	File	Line	Snippet	Context
LOW⚡	openrouter/openrouter_api_server.py	963	# Step 1: Run LM sample generation	COMMENT
LOW	openrouter/openrouter_api_server.py	992	# Step 2: Run audio generation with heartbeats	COMMENT
LOW⚡	acestep/llm_inference_cfg_fixes_test.py	438	# Step 0: output EOS; step 1+: output token 5	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	719	# Step 1: Encode full string "keyscale: {keyscale}" (with space, as tokenizer sees it)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	723	# Step 2: Find where context ends in full_token_ids	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	738	# Step 3: Extract keyscale value tokens (everything after context)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	741	# Step 4: Verify we extracted some tokens (sanity check)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	747	# Step 5: Verify first token is a note (A-G)	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	759	# Step 6: Build prefix mappings from keyscale value tokens	COMMENT
LOW⚡	acestep/ui/gradio/i18n/en.json	528	"training_dataset": "## Dataset Builder Tutorial\n\n### Step 1: Load or Scan\n- Load: Enter path to existing dat	CODE
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/turbo/modeling_acestep_v15_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/sft/modeling_acestep_v15_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	329	# Step 1: Self-attention with AdaLN	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	339	# Step 2: Cross-attention	COMMENT
LOW⚡	acestep/models/mlx/dit_model.py	350	# Step 3: MLP with AdaLN	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	159	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	165	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	168	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	acestep/models/base/modeling_acestep_v15_base.py	171	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	503	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	519	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	534	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW	acestep/training_v2/ui/flows_train.py	179	# Step 0: config depth	COMMENT

Modern Structural Boilerplate58 hits · 62 pts

Severity	File	Line	Snippet	Context
LOW	acestep/ui/gradio/__init__.py	8	__all__ = ["create_gradio_interface"]	CODE
LOW	acestep/ui/gradio/events/results_handlers.py	75	__all__ = [	CODE
LOW	acestep/ui/gradio/events/wiring/__init__.py	32	__all__ = [	CODE
LOW	acestep/ui/gradio/events/results/batch_management.py	22	__all__ = [	CODE
LOW	acestep/ui/gradio/events/generation/llm_actions.py	15	__all__ = [	CODE
LOW	acestep/core/scoring/__init__.py	13	__all__ = [	CODE
LOW	acestep/core/lora/__init__.py	8	__all__ = ["collect_adapter_names", "build_lora_registry", "apply_scale_to_adapter", "LoraService"]	CODE
LOW	acestep/core/generation/handler/__init__.py	41	__all__ = [	CODE
LOW	acestep/training/__init__.py	55	__all__ = [	CODE
LOW	acestep/training/dataset_builder.py	9	__all__ = [	CODE
LOW	acestep/training/path_safety.py	38	def set_safe_root(root: str) -> None:	CODE
LOW	acestep/training/dataset_builder_modules/__init__.py	4	__all__ = [	CODE
LOW	acestep/models/turbo/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/xl_turbo/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/xl_sft/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/xl_sft/apg_guidance.py	15	__all__ = [	CODE
LOW	acestep/models/common/configuration_acestep_v15.py	265	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/common/dcw_correction.py	44	__all__ = [	CODE
LOW	acestep/models/xl_base/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/xl_base/apg_guidance.py	15	__all__ = [	CODE
LOW	acestep/models/sft/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/sft/apg_guidance.py	15	__all__ = [	CODE
LOW	acestep/models/mlx/dit_convert.py	8	logger = logging.getLogger(__name__)	CODE
LOW	acestep/models/mlx/__init__.py	7	logger = logging.getLogger(__name__)	CODE
LOW	acestep/models/mlx/vae_convert.py	15	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/models/mlx/vae_model.py	17	logger = logging.getLogger(__name__)	CODE
LOW	acestep/models/mlx/dcw_correction_mlx.py	28	logger = logging.getLogger(__name__)	CODE
LOW	acestep/models/base/configuration_acestep_v15.py	5	__all__ = ["AceStepConfig"]	CODE
LOW	acestep/models/base/apg_guidance.py	15	__all__ = [	CODE
LOW	acestep/api/job_llm_preparation.py	6	__all__ = [	CODE
LOW	…ep/api/train_api_dataset_scan_load_routes_http_test.py	81	def set_all_instrumental(self, value: bool) -> None:	CODE
LOW	…ep/api/train_api_dataset_scan_load_routes_http_test.py	86	def set_custom_tag(self, tag: str, position: str) -> None:	CODE
LOW	acestep/api/http/auth.py	12	def set_api_key(key: Optional[str]) -> None:	CODE
LOW	acestep/api/jobs/store.py	148	def update_progress(self, job_id: str, progress: float, stage: Optional[str] = None) -> None:	CODE
LOW	acestep/api/jobs/store.py	194	def update_status_text(self, job_id: str, text: str) -> None:	CODE
LOW	acestep/api/jobs/store.py	200	def update_progress_text(self, job_id: str, text: str) -> None:	CODE
LOW	acestep/training_v2/preprocess_discovery.py	18	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/trainer_helpers.py	29	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/preprocess.py	39	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/trainer_basic_loop.py	26	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/trainer_fixed.py	55	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/model_discovery.py	18	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/estimate.py	23	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/settings.py	23	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/gpu_utils.py	18	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/model_loader.py	22	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/tensorboard_utils.py	20	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/fixed_lora_module.py	67	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/trainer_vanilla.py	58	logger = logging.getLogger(__name__)	CODE
LOW⚡	acestep/training_v2/optim.py	36	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/ui/presets.py	29	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/ui/__init__.py	42	def set_plain_mode(value: bool) -> None:	CODE
LOW⚡	acestep/training_v2/ui/flows_setup.py	27	logger = logging.getLogger(__name__)	CODE
LOW	acestep/training_v2/ui/flows.py	13	__all__ = ["wizard_train", "wizard_preprocess", "wizard_estimate"]	CODE
LOW	acestep/training_v2/cli/config_builder.py	20	logger = logging.getLogger(__name__)	CODE
LOW	acestep/text_tasks/passphrase_store_test.py	56	def set_password(service: str, username: str, passphrase: str) -> None:	CODE
LOW	acestep/text_tasks/secure_secret_store_test.py	71	def set_password(service: str, username: str, secret: str) -> None:	CODE
LOW	acestep/text_tasks/secure_secret_store_test.py	155	def set_password(*_args) -> None:	CODE

Over-Commented Block66 hits · 52 pts

Severity	File	Line	Snippet	Context
LOW	install_uv.sh	1	#!/usr/bin/env bash	COMMENT
LOW	docker-compose.jetson.yml	1	# =============================================================================	COMMENT
LOW	docker-compose.jetson.yml	21	#	COMMENT
LOW	docker-compose.jetson.yml	61	# acestep-v15-base — Full base model (Extract/Lego/Complete)	COMMENT
LOW	start_api_server.sh	21	# Preferred download source: auto (default), huggingface, or modelscope	COMMENT
LOW	requirements-rocm.txt	1	# ACE-Step 1.5 - AMD ROCm 7.2 Windows Dependencies	COMMENT
LOW	start_gradio_ui_rocm.sh	1	#!/usr/bin/env bash	COMMENT
LOW	start_api_server_rocm.sh	1	#!/usr/bin/env bash	COMMENT
LOW	start_api_server_rocm.sh	41	# API key for authentication (optional)	COMMENT
LOW	start_gradio_ui.sh	101	# LLM (Language Model) initialization settings	COMMENT
LOW	docker-compose.yml	1	# =============================================================================	COMMENT
LOW	start_gradio_ui_rocm_manual.sh	1	#!/usr/bin/env bash	COMMENT
LOW	start_gradio_ui_manual.sh	221		COMMENT
LOW	start_api_server_macos.sh	1	#!/usr/bin/env bash	COMMENT
LOW	start_api_server_macos.sh	21	PORT=8001	COMMENT
LOW	requirements-sidestep.txt	1	# ╔══════════════════════════════════════════════════════════════╗	COMMENT
LOW	requirements-sidestep.txt	21	# Uncomment to enable the Prodigy optimizer.	COMMENT
LOW	acestep/constants.py	61	# Valid time signatures - common musical meter patterns	COMMENT
LOW	acestep/constants.py	81	# - cover: Style transfer using reference audio	COMMENT
LOW	acestep/constants.py	141		COMMENT
LOW	acestep/inference.py	581	need_audio_codes = not user_provided_audio_codes	COMMENT
LOW	acestep/inference.py	621	use_random_seed_for_dit = False	COMMENT
LOW	acestep/debug_utils.py	21		COMMENT
LOW	acestep/third_parts/nano-vllm/nanovllm/utils/context.py	21	# `_CONTEXT` variable. In concurrent serving scenarios (API server with	COMMENT
LOW	…/core/generation/handler/service_generate_flow_edit.py	81	src_text_hs, src_lyric_hs = embed_source(handler, src_text_ids, src_lyric_ids)	COMMENT
LOW	acestep/models/__init__.py	1	# ACE-Step model definitions	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	101	# Causal sliding: only attend to past window steps	COMMENT
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	801	# 如果有 padding mask (attention_mask [B, L])，直接传给它即可。	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	101	# Causal sliding: only attend to past window steps	COMMENT
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	801	# 如果有 padding mask (attention_mask [B, L])，直接传给它即可。	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	101	# (B) Handle sliding window	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	641		COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	801	# 场景 A: Flash Attention 模式	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1121	# 如果有 padding mask (attention_mask [B, L])，直接传给它即可。	COMMENT
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1401	# 判断是否使用 Flash Attention 2	COMMENT
LOW	acestep/models/common/configuration_acestep_v15.py	1	# coding=utf-8	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	101	# (B) Handle sliding window	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	641		COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	801	# 场景 A: Flash Attention 模式	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1121	# 如果有 padding mask (attention_mask [B, L])，直接传给它即可。	COMMENT
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1401	# 判断是否使用 Flash Attention 2	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	101	# (B) Handle sliding window	COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	641		COMMENT
LOW	acestep/models/sft/modeling_acestep_v15_base.py	801	# 场景 A: Flash Attention 模式	COMMENT
LOW	acestep/models/mlx/vae_model.py	41	# weights (exp overflows float16 at alpha > ~11). This is only a problem	COMMENT
LOW	acestep/models/mlx/dit_generate.py	1	# MLX diffusion generation loop for AceStep DiT decoder.	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	1	# Copyright 2025 The ACESTEO Team. All rights reserved.	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	101	# (B) Handle sliding window	COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	641		COMMENT
LOW	acestep/models/base/modeling_acestep_v15_base.py	801	# 场景 A: Flash Attention 模式	COMMENT
LOW	…-transcription/scripts/acestep-lyrics-transcription.sh	1	#!/bin/bash	COMMENT
LOW	.claude/skills/acestep/scripts/acestep.sh	1	#!/bin/bash	COMMENT
LOW	…/skills/acestep-thumbnail/scripts/acestep-thumbnail.sh	1	#!/bin/bash	COMMENT
LOW	.claude/skills/acestep-simplemv/scripts/render-mv.sh	1	#!/bin/bash	COMMENT
LOW	.claude/skills/acestep-simplemv/scripts/render.sh	1	#!/bin/bash	COMMENT
6 more matches not shown…

AI Structural Patterns49 hits · 33 pts

Severity	File	Line	Context
LOW	openrouter/openrouter_api_server.py	1124	CODE
LOW	acestep/llm_inference.py	874	CODE
LOW	acestep/llm_inference.py	1145	CODE
LOW	acestep/llm_inference.py	1282	CODE
LOW	acestep/llm_inference.py	2042	CODE
LOW	acestep/llm_inference.py	3960	CODE
LOW	acestep/inference.py	1290	CODE
LOW	acestep/ui/gradio/events/results/batch_queue.py	14	CODE
LOW	acestep/ui/gradio/events/results/generation_progress.py	42	CODE
LOW	acestep/third_parts/nano-vllm/nanovllm/models/qwen3.py	88	CODE
LOW	acestep/core/generation/handler/service_generate.py	21	CODE
LOW	…step/core/generation/handler/generate_music_request.py	222	CODE
LOW	acestep/core/generation/handler/diffusion.py	18	CODE
LOW	…step/core/generation/handler/generate_music_execute.py	18	CODE
LOW	…p/core/generation/handler/init_service_orchestrator.py	48	CODE
LOW	…ep/core/generation/handler/service_generate_execute.py	72	CODE
LOW	acestep/core/generation/handler/conditioning_batch.py	21	CODE
LOW	acestep/core/generation/handler/generate_music.py	195	CODE
LOW	…tep/core/generation/handler/conditioning_masks_test.py	25	CODE
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	478	CODE
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	1307	CODE
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	1701	CODE
LOW	acestep/models/turbo/modeling_acestep_v15_turbo.py	1852	CODE
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	478	CODE
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	1323	CODE
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	1713	CODE
LOW	…estep/models/xl_turbo/modeling_acestep_v15_xl_turbo.py	1864	CODE
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	481	CODE
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1326	CODE
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1716	CODE
LOW	acestep/models/xl_sft/modeling_acestep_v15_xl_base.py	1848	CODE
LOW	acestep/models/common/configuration_acestep_v15.py	148	CODE
LOW	acestep/models/common/flow_edit_pipeline.py	76	CODE
LOW	acestep/models/common/flow_edit.py	54	CODE
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	481	CODE
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1326	CODE
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1716	CODE
LOW	acestep/models/xl_base/modeling_acestep_v15_xl_base.py	1848	CODE
LOW	acestep/models/sft/modeling_acestep_v15_base.py	481	CODE
LOW	acestep/models/sft/modeling_acestep_v15_base.py	1310	CODE
LOW	acestep/models/sft/modeling_acestep_v15_base.py	1704	CODE
LOW	acestep/models/sft/modeling_acestep_v15_base.py	1836	CODE
LOW	acestep/models/mlx/dit_model.py	432	CODE
LOW	acestep/models/mlx/dit_generate.py	168	CODE
LOW	acestep/models/base/modeling_acestep_v15_base.py	481	CODE
LOW	acestep/models/base/modeling_acestep_v15_base.py	1310	CODE
LOW	acestep/models/base/modeling_acestep_v15_base.py	1704	CODE
LOW	acestep/models/base/modeling_acestep_v15_base.py	1836	CODE
LOW	acestep/training_v2/ui/help_formatter.py	94	CODE

Self-Referential Comments9 hits · 30 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	openrouter/client_test.py	22	# This file is an executable API client script, not a pytest test module.	COMMENT
MEDIUM	acestep/audio_utils.py	83	# Create a copy to avoid modifying original in-place	COMMENT
MEDIUM	acestep/inference.py	965	# Create a copy of params dict for this audio	COMMENT
MEDIUM	acestep/ui/gradio/events/training/preprocess_test.py	34	# Create a dummy .pt file	COMMENT
MEDIUM	acestep/ui/gradio/events/results/generation_info.py	16	# This file is in acestep/ui/gradio/events/results/, need 6 levels up to project root	COMMENT
MEDIUM	acestep/ui/gradio/events/generation/metadata_loading.py	160	# This file is in acestep/ui/gradio/events/generation/	COMMENT
MEDIUM	acestep/ui/gradio/events/generation/service_init.py	123	# This file is in acestep/ui/gradio/events/generation/	COMMENT
MEDIUM⚡	acestep/training/data_module.py	26	# Create a dummy class for type hints	COMMENT
MEDIUM	acestep/training/data_module_test.py	66	# Create a manifest with one good and one bad path	COMMENT

Example Usage Blocks8 hits · 14 pts

Severity	File	Line	Snippet	Context
LOW	install_uv.sh	5	# Usage:	COMMENT
LOW⚡	docker-compose.jetson.yml	9	# Usage:	COMMENT
LOW⚡	docker-compose.yml	5	# Usage:	COMMENT
LOW	…-transcription/scripts/acestep-lyrics-transcription.sh	7	# Usage:	COMMENT
LOW	.claude/skills/acestep/scripts/acestep.sh	7	# Usage:	COMMENT
LOW	…/skills/acestep-thumbnail/scripts/acestep-thumbnail.sh	4	# Usage:	COMMENT
LOW	.claude/skills/acestep-simplemv/scripts/render-mv.sh	4	# Usage:	COMMENT
LOW	.claude/skills/acestep-simplemv/scripts/render.sh	4	# Usage:	COMMENT

AI Slop Vocabulary5 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW⚡	acestep/constrained_logits_processor.py	1793	# Avoid infinite recursion: if we're still in a fixed_strings state, just return scores	COMMENT
LOW⚡	acestep/constrained_logits_processor.py	1795	# This shouldn't happen, but if it does, just return scores to avoid recursion	COMMENT
MEDIUM⚡	acestep/ui/gradio/i18n/en.json	518	"generation_repaint": "## Repaint Mode Tutorial\n\nBest for: Fixing or regenerating specific sections of generat	CODE
LOW	…ep/third_parts/nano-vllm/nanovllm/engine/llm_engine.py	65	# For now, just use the same prompt (user should provide unconditional_prompt)	COMMENT
MEDIUM	acestep/core/scoring/lm_score.py	339	# Calculate recall using the robust logic	COMMENT

Modern AI Meta-Vocabulary3 hits · 10 pts

Severity	File	Line	Snippet	Context
MEDIUM	acestep/ui/gradio/i18n/pt.json	513	"generation_edit": "## Edit (Overlay Flow-Edit)\n\nIdeal para: transformar um áudio existente em direção a uma n	CODE
MEDIUM⚡	acestep/ui/gradio/i18n/en.json	526	"generation_edit": "## Edit (Flow-Edit Overlay)\n\nBest for: morphing an existing audio toward new lyrics or a n	CODE
MEDIUM⚡	acestep/training_v2/trainer_fixed.py	68	# FixedLoRATrainer -- orchestration	COMMENT

Slop Phrases2 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW⚡	acestep/llm_inference.py	2907	# Don't forget to save the last field	COMMENT
MEDIUM	.github/workflows/codeql.yml	54	# If you are analyzing a compiled language, you can modify the 'build-mode' for that language to customize how	COMMENT

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings