xorbitsai/inference

18.3

Adjusted Score

18.3

Raw Score

100%

Time Factor

2026-08-28

Last Push

9.5K

Stars

Python

Language

491.1K

Lines of Code

1.7K

Files

7.3K

Pattern Hits

2026-08-29

Scan Date

0.17

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

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

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 1HIGH 291MEDIUM 483LOW 6500

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 7275 distinct pattern matches across 25 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 Identifiers2799 hits · 2732 pts

Severity	File	Line	Snippet	Context
LOW	build_backend.py	43	def _validate_builtin_model_specs(root=None):	CODE
LOW	benchmark/utils.py	169	def _num_special_tokens_to_add(tokenizer: "PreTrainedTokenizerBase") -> int:	CODE
LOW	benchmark/utils.py	234	def _gen_prompt_decode_to_target_len(	CODE
LOW	benchmark/utils.py	300	def _generate_prompt_with_target_len(	CODE
LOW	benchmark/benchmark_serving.py	30	def _validate_random_range_ratio(value: float, name: str) -> float:	CODE
LOW	…register-model/[modelType]/[modelName]/page-client.tsx	9	export default function RegisterModelEditPageClient() {	CODE
LOW	…ntend/src/app/running-model/[modelUid]/page-client.tsx	7	export default function RunningModelDetailPageClient() {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	234	export function getLaunchModelEndpointType(type: RequestModelType) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	288	function restoreNestedFormListObject(values: FormValues, parent: string, field: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	303	function transformOnlyValueFormListToArray(values: FormValues, field: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	327	function transformArrayToOnlyValueFormList(values: FormValues, field: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	377	function applyNestedFormListObject(values: FormValues, parent: string, field: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	399	function deleteNestedEmptyArrayField(values: FormValues, parent: string, field: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	688	function parseReplicaConfigCommandValue(value: string) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	754	export function generateCommandLineStatement(params: FormValues) {	CODE
LOW	frontend/src/components/pages/launch-model/utils.tsx	1099	export function isVisibleRequiredLaunchField(field: LaunchFieldConfig) {	CODE
LOW	…nents/pages/running-model-detail/capability-config.tsx	144	function appendCommonMediaFormData(formData: FormData, context: TransformContext) {	CODE
LOW	…nents/pages/running-model-detail/capability-config.tsx	174	function appendCommonVideoFormData(formData: FormData, context: TransformContext) {	CODE
LOW	…end/src/components/pages/running-model-detail/utils.ts	205	export function transformFileInfoForResult(message?: ChatChoicesMessage) {	CODE
LOW	…nts/pages/running-model-detail/emotion-vector-utils.ts	28	export function parseIndexTTSEmotionVector(value: unknown): number[] \| undefined {	CODE
LOW	…ts/pages/running-model-detail/panels/result-panels.tsx	394	function SpeakerEmbeddingResultPanel({ result }: { result: unknown }) {	CODE
LOW	…ts/pages/token-router/router-runtime-detail-drawer.tsx	61	export function RouterRuntimeDetailDrawer({ router, open, onOpenChange }: Props) {	CODE
LOW	…ponents/pages/token-router/router-config-normalizer.ts	224	export function normalizeTokenRouterRuntimeInstances(value: unknown): TokenRouterRuntimeInstance[] {	CODE
LOW	…ponents/pages/token-router/router-config-normalizer.ts	230	export function normalizeTokenRouterRuntimeInstance(	CODE
LOW	…ponents/pages/token-router/router-config-normalizer.ts	271	export function normalizeTokenRouterAssignments(value: unknown): TokenRouterAssignment[] {	CODE
LOW	…s/pages/token-router/tokenizer-asset-binding-status.ts	13	export function getBindingDesiredStateLabel(t: TFunc, state: BindingDesiredState): string {	CODE
LOW	…s/pages/token-router/tokenizer-asset-binding-status.ts	17	export function getBindingObservedStateLabel(t: TFunc, state: string): string {	CODE
LOW	…s/pages/token-router/tokenizer-asset-binding-status.ts	43	export function getBindingObservedStateBadgeVariant(state: string): BadgeVariant {	CODE
LOW	xinference/conftest.py	165	def run_test_cluster_in_subprocess(	CODE
LOW	xinference/constants.py	122	def _parse_virtual_env_find_links_allowed_roots(	CODE
LOW	xinference/constants.py	186	def _get_or_create_persisted_secret(env_name: str, file_name: str) -> str:	CODE
LOW	xinference/types.py	369	def get_pydantic_model_from_method(	CODE
LOW⚡	xinference/device_utils.py	53	def _count_with_visible_devices(env_name: str, device_count: int) -> int:	CODE
LOW⚡	xinference/device_utils.py	273	def _collect_gpu_info_with_env(	CODE
LOW⚡	xinference/device_utils.py	583	def move_model_to_available_device(model):	CODE
LOW⚡	xinference/device_utils.py	590	def get_device_preferred_dtype(device: str) -> Union[torch.dtype, None]:	CODE
LOW⚡	xinference/device_utils.py	599	def is_hf_accelerate_supported(device: str) -> bool:	CODE
LOW⚡	xinference/device_utils.py	609	def get_available_device_env_name():	CODE
LOW	xinference/device_utils.py	635	def get_per_process_gpu_memory() -> Dict[int, Dict[int, int]]:	CODE
LOW⚡	xinference/core/worker.py	391	def _snapshot_gpu_occupying_pids(device_indices: list) -> set:	CODE
LOW⚡	xinference/core/worker.py	484	def _process_or_ancestor_has_uid(proc: Any, uid_lower: str) -> bool:	CODE
LOW⚡	xinference/core/worker.py	1722	def _load_persisted_launch_args(self) -> dict:	CODE
LOW	xinference/core/worker.py	242	def _wait_for_metrics_export_server(	CODE
LOW	xinference/core/worker.py	259	def _exclusive_venv_path_lock(env_path: str):	CODE
LOW	xinference/core/worker.py	595	async def _kill_gpu_orphans_by_ppid(	CODE
LOW	xinference/core/worker.py	997	async def _evict_replica_from_supervisor(self, model_uid: str) -> None:	CODE
LOW	xinference/core/worker.py	1036	def _get_spec_dicts_with_cache_status(	CODE
LOW	xinference/core/worker.py	1108	async def _cleanup_gpu_orphans_on_startup(self) -> None:	CODE
LOW	xinference/core/worker.py	1386	async def _get_supervisor_ref_with_generation(	CODE
LOW	xinference/core/worker.py	1410	async def _get_supervisor_ref_on_actor_loop(	CODE
LOW	xinference/core/worker.py	1526	async def _record_model_versions_best_effort(self, cache_tracker_ref):	CODE
LOW	xinference/core/worker.py	1562	def _clear_unregistered_supervisor_refs(self) -> bool:	CODE
LOW	xinference/core/worker.py	1578	def _get_supervisor_ref_generation(	CODE
LOW	xinference/core/worker.py	1615	async def _refresh_supervisor_address(self):	CODE
LOW	xinference/core/worker.py	1632	def _get_running_replica_states(self) -> List[Dict[str, Any]]:	CODE
LOW	xinference/core/worker.py	1705	def _remove_persisted_launch_args(self, model_uid: str):	CODE
LOW	xinference/core/worker.py	1913	async def allocate_devices_with_gpu_idx(	CODE
LOW	xinference/core/worker.py	1951	async def get_gpu_allocation_status(self) -> Dict[str, Any]:	CODE
LOW	xinference/core/worker.py	1999	async def _append_sub_pool_protected(	CODE
LOW	xinference/core/worker.py	2063	async def _allocate_subpool_devices(	CODE
2739 more matches not shown…

Cross-File Repetition228 hits · 1140 pts

Severity	File	Line	Snippet	Context
HIGH	xinference/model/video/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/llm/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/embedding/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/image/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/audio/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/rerank/__init__.py	0	load downloaded json configurations from the builtin directory.	STRING
HIGH	xinference/model/video/engine_family.py	0	whether the engine implements the model, ignoring host constraints.	STRING
HIGH	xinference/model/audio/engine_family.py	0	whether the engine implements the model, ignoring host constraints.	STRING
HIGH	xinference/model/world/engine_family.py	0	whether the engine implements the model, ignoring host constraints.	STRING
HIGH	…ence/model/llm/tool_parsers/deepseek_v3_tool_parser.py	0	extract tool calls from complete model output. parses the model output to find json code blocks containing tool calls an	STRING
HIGH	xinference/model/llm/tool_parsers/glm5_tool_parser.py	0	extract tool calls from complete model output. parses the model output to find json code blocks containing tool calls an	STRING
HIGH	…ce/model/llm/tool_parsers/deepseek_v3_1_tool_parser.py	0	extract tool calls from complete model output. parses the model output to find json code blocks containing tool calls an	STRING
HIGH	xinference/model/llm/tool_parsers/glm4_tool_parser.py	0	extract tool calls from complete model output. parses the model output to find json code blocks containing tool calls an	STRING
HIGH	xinference/thirdparty/cosyvoice/tokenizer/tokenizer.py	0	's\|'t\|'re\|'ve\|'m\|'ll\|'d\| ?\p{l}+\| ?\p{n}+\| ?[^\s\p{l}\p{n}]+\|\s+(?!\s)\|\s+	STRING
HIGH	xinference/thirdparty/indextts/utils/tokenizer.py	0	's\|'t\|'re\|'ve\|'m\|'ll\|'d\| ?\p{l}+\| ?\p{n}+\| ?[^\s\p{l}\p{n}]+\|\s+(?!\s)\|\s+	STRING
HIGH	xinference/thirdparty/whisper/tokenizer.py	0	's\|'t\|'re\|'ve\|'m\|'ll\|'d\| ?\p{l}+\| ?\p{n}+\| ?[^\s\p{l}\p{n}]+\|\s+(?!\s)\|\s+	STRING
HIGH	…ference/thirdparty/cosyvoice/transformer/activation.py	0	implementation of a sine-based periodic activation function shape: - input: (b, c, t) - output: (b, c, t), same shape as	STRING
HIGH	xinference/thirdparty/indextts/BigVGAN/activations.py	0	implementation of a sine-based periodic activation function shape: - input: (b, c, t) - output: (b, c, t), same shape as	STRING
HIGH	…thirdparty/indextts/s2mel/modules/hifigan/generator.py	0	implementation of a sine-based periodic activation function shape: - input: (b, c, t) - output: (b, c, t), same shape as	STRING
HIGH	…irdparty/indextts/s2mel/modules/bigvgan/activations.py	0	implementation of a sine-based periodic activation function shape: - input: (b, c, t) - output: (b, c, t), same shape as	STRING
HIGH	…ference/thirdparty/cosyvoice/transformer/activation.py	0	initialization. input: - in_features: shape of the input - alpha: trainable parameter alpha is initialized to 1 by defau	STRING
HIGH	xinference/thirdparty/indextts/BigVGAN/activations.py	0	initialization. input: - in_features: shape of the input - alpha: trainable parameter alpha is initialized to 1 by defau	STRING
HIGH	…thirdparty/indextts/s2mel/modules/hifigan/generator.py	0	initialization. input: - in_features: shape of the input - alpha: trainable parameter alpha is initialized to 1 by defau	STRING
HIGH	…irdparty/indextts/s2mel/modules/bigvgan/activations.py	0	initialization. input: - in_features: shape of the input - alpha: trainable parameter alpha is initialized to 1 by defau	STRING
HIGH	…ference/thirdparty/cosyvoice/transformer/activation.py	0	forward pass of the function. applies the function to the input elementwise. snake ∶= x + 1/a * sin^2 (xa)	STRING
HIGH	xinference/thirdparty/indextts/BigVGAN/activations.py	0	forward pass of the function. applies the function to the input elementwise. snake ∶= x + 1/a * sin^2 (xa)	STRING
HIGH	…thirdparty/indextts/s2mel/modules/hifigan/generator.py	0	forward pass of the function. applies the function to the input elementwise. snake ∶= x + 1/a * sin^2 (xa)	STRING
HIGH	…irdparty/indextts/s2mel/modules/bigvgan/activations.py	0	forward pass of the function. applies the function to the input elementwise. snake ∶= x + 1/a * sin^2 (xa)	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	0	make mask tensor containing indices of padded part. see description of make_non_pad_mask. args: lengths (torch.tensor):	STRING
HIGH	xinference/thirdparty/indextts/utils/common.py	0	make mask tensor containing indices of padded part. see description of make_non_pad_mask. args: lengths (torch.tensor):	STRING
HIGH	xinference/thirdparty/indextts/utils/utils.py	0	make mask tensor containing indices of padded part. see description of make_non_pad_mask. args: lengths (torch.tensor):	STRING
HIGH	…ference/thirdparty/deepseek_vl2/models/conversation.py	0	a class that manages prompt templates and keeps all conversation history.	STRING
HIGH	…irdparty/sensenova_u1/models/neo_unify/conversation.py	0	a class that manages prompt templates and keeps all conversation history.	STRING
HIGH	xinference/thirdparty/deepseek_vl/utils/conversation.py	0	a class that manages prompt templates and keeps all conversation history.	STRING
HIGH	xinference/thirdparty/internvl/conversation.py	0	a class that manages prompt templates and keeps all conversation history.	STRING
HIGH	…ference/thirdparty/deepseek_vl2/models/conversation.py	0	update the last output. the last message is typically set to be none when constructing the prompt, so we need to update	STRING
HIGH	…irdparty/sensenova_u1/models/neo_unify/conversation.py	0	update the last output. the last message is typically set to be none when constructing the prompt, so we need to update	STRING
HIGH	xinference/thirdparty/deepseek_vl/utils/conversation.py	0	update the last output. the last message is typically set to be none when constructing the prompt, so we need to update	STRING
HIGH	xinference/thirdparty/internvl/conversation.py	0	update the last output. the last message is typically set to be none when constructing the prompt, so we need to update	STRING
HIGH	…ference/thirdparty/deepseek_vl2/models/conversation.py	0	convert the conversation to openai chat completion format.	STRING
HIGH	…irdparty/sensenova_u1/models/neo_unify/conversation.py	0	convert the conversation to openai chat completion format.	STRING
HIGH	xinference/thirdparty/deepseek_vl/utils/conversation.py	0	convert the conversation to openai chat completion format.	STRING
HIGH	xinference/thirdparty/internvl/conversation.py	0	convert the conversation to openai chat completion format.	STRING
HIGH	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	0	this is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). the hidden states go from (batch, num_key_va	STRING
HIGH	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	0	this is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). the hidden states go from (batch, num_key_va	STRING
HIGH	…ference/thirdparty/hidream_o1/qwen3_vl_transformers.py	0	this is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). the hidden states go from (batch, num_key_va	STRING
HIGH	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	0	multi-headed attention from 'attention is all you need' paper	STRING
HIGH	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	0	multi-headed attention from 'attention is all you need' paper	STRING
HIGH	…ference/thirdparty/hidream_o1/qwen3_vl_transformers.py	0	multi-headed attention from 'attention is all you need' paper	STRING
HIGH	…ish_speech/fish_speech/models/vqgan/modules/firefly.py	0	remove padding from x, handling properly zero padding. only for 1d!	STRING
HIGH	…s/maskgct/models/codec/speechtokenizer/modules/conv.py	0	remove padding from x, handling properly zero padding. only for 1d!	STRING
HIGH	…ference/thirdparty/indextts/s2mel/dac/model/encodec.py	0	remove padding from x, handling properly zero padding. only for 1d!	STRING
HIGH	xinference/thirdparty/indextts/s2mel/modules/encodec.py	0	remove padding from x, handling properly zero padding. only for 1d!	STRING
HIGH	…ish_speech/fish_speech/models/vqgan/modules/firefly.py	0	tiny wrapper around f.pad, just to allow for reflect padding on small input. if this is the case, we insert extra 0 padd	STRING
HIGH	…s/maskgct/models/codec/speechtokenizer/modules/conv.py	0	tiny wrapper around f.pad, just to allow for reflect padding on small input. if this is the case, we insert extra 0 padd	STRING
HIGH	…ference/thirdparty/indextts/s2mel/dac/model/encodec.py	0	tiny wrapper around f.pad, just to allow for reflect padding on small input. if this is the case, we insert extra 0 padd	STRING
HIGH	xinference/thirdparty/indextts/s2mel/modules/encodec.py	0	tiny wrapper around f.pad, just to allow for reflect padding on small input. if this is the case, we insert extra 0 padd	STRING
HIGH	…y/sensenova_u1/models/neo_unify/modeling_fm_modules.py	0	embeds scalar timesteps into vector representations.	STRING
HIGH	…ference/thirdparty/hidream_o1/qwen3_vl_transformers.py	0	embeds scalar timesteps into vector representations.	STRING
HIGH	…dparty/indextts/s2mel/modules/diffusion_transformer.py	0	embeds scalar timesteps into vector representations.	STRING
168 more matches not shown…

Unused Imports1024 hits · 976 pts

Severity	File	Line	Context
LOW	build_backend.py	35	CODE
LOW	benchmark/utils.py	29	CODE
LOW	xinference/__init__.py	37	CODE
LOW	xinference/device_utils.py	131	CODE
LOW	xinference/device_utils.py	140	CODE
LOW	xinference/device_utils.py	149	CODE
LOW	xinference/device_utils.py	158	CODE
LOW	xinference/device_utils.py	159	CODE
LOW	xinference/device_utils.py	168	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	22	CODE
LOW	xinference/_compat.py	34	CODE
LOW	xinference/_compat.py	35	CODE
LOW	xinference/_compat.py	36	CODE
LOW	xinference/_compat.py	37	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	39	CODE
LOW	xinference/_compat.py	51	CODE
LOW	xinference/_compat.py	52	CODE
LOW	xinference/_compat.py	53	CODE
LOW	xinference/_compat.py	54	CODE
LOW	xinference/core/tokenizer_asset_store.py	4	CODE
LOW	xinference/core/router_orchestration.py	4	CODE
LOW	xinference/core/router_config_store.py	10	CODE
LOW	xinference/core/supervisor.py	64	CODE
LOW	xinference/core/supervisor.py	90	CODE
LOW	xinference/core/supervisor.py	91	CODE
LOW	xinference/core/supervisor.py	93	CODE
LOW	xinference/core/supervisor.py	94	CODE
LOW	xinference/core/supervisor.py	95	CODE
LOW	xinference/core/supervisor.py	96	CODE
LOW	xinference/core/launch_strategy.py	22	CODE
LOW	xinference/core/router_scheduler.py	4	CODE
LOW	xinference/core/router_node_store.py	4	CODE
LOW	xinference/core/http_protocol.py	131	CODE
LOW	xinference/core/router_assignment_store.py	4	CODE
LOW	xinference/core/model.py	49	CODE
LOW	xinference/core/model.py	50	CODE
LOW	xinference/core/tokenizer_asset_registry.py	4	CODE
LOW	xinference/core/router_deployment_store.py	4	CODE
LOW	xinference/core/system_settings_store.py	17	CODE
LOW	xinference/core/router_registry.py	4	CODE
LOW	xinference/core/tests/test_monitoring_v2.py	1	CODE
LOW	xinference/core/tests/test_router_orchestration.py	3	CODE
LOW	xinference/core/tests/test_tokenizer_asset_store.py	3	CODE
LOW	xinference/deploy/cmdline.py	57	CODE
LOW	xinference/deploy/router_agent.py	4	CODE
LOW	xinference/deploy/router_agent.py	6	CODE
LOW	xinference/deploy/utils.py	43	CODE
LOW	xinference/deploy/router.py	4	CODE
964 more matches not shown…

Decorative Section Separators230 hits · 762 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	xinference/device_utils.py	29	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/device_utils.py	31	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/device_utils.py	48	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/device_utils.py	50	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/device_utils.py	268	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/device_utils.py	270	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	90	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	92	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	175	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	177	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	449	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	451	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	558	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/device_utils.py	560	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/metrics.py	37	# ===========================================================================	COMMENT
MEDIUM⚡	xinference/core/metrics.py	39	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	58	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	60	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	102	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	104	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	338	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	340	# ===========================================================================	COMMENT
MEDIUM	xinference/core/metrics.py	367	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/metrics.py	369	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/metrics.py	434	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/metrics.py	436	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/metrics.py	450	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/metrics.py	452	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/otel.py	161	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/otel.py	163	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/otel.py	316	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/otel.py	318	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/tests/test_model.py	302	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	xinference/core/tests/test_model.py	308	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_utils.py	682	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_utils.py	684	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	114	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	116	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	153	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	155	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	187	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	189	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	211	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	213	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	253	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	258	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	356	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	358	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	417	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	419	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	505	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	xinference/core/tests/test_subpool_hardening.py	507	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	357	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	359	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	394	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	396	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	458	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	460	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	498	# ------------------------------------------------------------------	COMMENT
MEDIUM	…nference/deploy/docker/pypiserver/download_packages.py	502	# ------------------------------------------------------------------	COMMENT
170 more matches not shown…

Excessive Try-Catch Wrapping756 hits · 713 pts

Severity	File	Line	Snippet	Context
LOW	build_backend.py	95	except Exception:	CODE
LOW	benchmark/benchmark_runner.py	198	except Exception:	CODE
MEDIUM	xinference/conftest.py	114	print(f"Error while checking endpoint: {e}")	CODE
LOW	xinference/__init__.py	31	except Exception:	CODE
LOW	xinference/__init__.py	38	except Exception:	CODE
MEDIUM	xinference/device_utils.py	61	def _mps_empty_cache():	CODE
MEDIUM	xinference/device_utils.py	313	def get_nvidia_gpu_info() -> Dict[str, Any]:	CODE
MEDIUM	xinference/device_utils.py	345	def get_metax_gpu_info() -> Dict:	CODE
MEDIUM	xinference/device_utils.py	360	def get_rocm_gpu_info() -> Dict:	CODE
LOW	xinference/device_utils.py	302	except Exception as e:	CODE
LOW	xinference/device_utils.py	309	except Exception:	CODE
LOW	xinference/device_utils.py	326	except Exception as e:	CODE
LOW	xinference/device_utils.py	340	except Exception as e:	CODE
LOW	xinference/device_utils.py	355	except Exception as e:	CODE
LOW	xinference/device_utils.py	369	except Exception:	CODE
LOW	xinference/device_utils.py	655	except Exception:	CODE
LOW	xinference/device_utils.py	678	except Exception:	CODE
MEDIUM	xinference/core/worker.py	1352	def signal_handler():	CODE
MEDIUM	xinference/core/worker.py	1369	def trigger_exit(self) -> bool:	CODE
MEDIUM	xinference/core/worker.py	1526	def _record_model_versions_best_effort(self, cache_tracker_ref):	CODE
LOW⚡	xinference/core/worker.py	386	except Exception as e:	CODE
LOW⚡	xinference/core/worker.py	401	except Exception:	CODE
LOW⚡	xinference/core/worker.py	411	except Exception:	CODE
LOW⚡	xinference/core/worker.py	416	except Exception:	CODE
LOW⚡	xinference/core/worker.py	477	except Exception:	CODE
LOW⚡	xinference/core/worker.py	493	except Exception:	CODE
LOW⚡	xinference/core/worker.py	1719	except Exception:	CODE
LOW⚡	xinference/core/worker.py	1730	except Exception:	CODE
LOW	xinference/core/worker.py	430	except Exception:	CODE
LOW	xinference/core/worker.py	441	except Exception:	CODE
LOW	xinference/core/worker.py	446	except Exception:	CODE
LOW	xinference/core/worker.py	543	except Exception:	CODE
LOW	xinference/core/worker.py	572	except Exception:	CODE
LOW	xinference/core/worker.py	587	except Exception:	CODE
LOW	xinference/core/worker.py	669	except Exception:	CODE
LOW	xinference/core/worker.py	804	except Exception:	CODE
LOW	xinference/core/worker.py	833	except Exception as e:	CODE
LOW	xinference/core/worker.py	858	except Exception:	CODE
LOW	xinference/core/worker.py	877	except Exception:	CODE
LOW	xinference/core/worker.py	879	except Exception:	CODE
LOW	xinference/core/worker.py	920	except Exception:	CODE
LOW	xinference/core/worker.py	945	except Exception as e:	CODE
LOW	xinference/core/worker.py	963	except Exception:	CODE
LOW	xinference/core/worker.py	987	except Exception:	CODE
LOW	xinference/core/worker.py	1019	except Exception:	CODE
LOW	xinference/core/worker.py	1153	except Exception:	CODE
LOW	xinference/core/worker.py	1155	except Exception:	CODE
LOW	xinference/core/worker.py	1196	except Exception:	CODE
LOW	xinference/core/worker.py	1308	except Exception:	CODE
LOW	xinference/core/worker.py	1316	except Exception:	CODE
LOW	xinference/core/worker.py	1356	except Exception as e:	CODE
LOW	xinference/core/worker.py	1372	except Exception as e:	CODE
LOW	xinference/core/worker.py	1440	except Exception:	CODE
LOW	xinference/core/worker.py	1555	except Exception:	CODE
LOW	xinference/core/worker.py	1699	except Exception:	CODE
LOW	xinference/core/worker.py	1765	except Exception:	CODE
LOW	xinference/core/worker.py	1777	except Exception:	CODE
LOW	xinference/core/worker.py	1816	except Exception:	CODE
LOW	xinference/core/worker.py	1841	except Exception:	CODE
LOW	xinference/core/worker.py	2056	except Exception:	CODE
696 more matches not shown…

Over-Commented Block601 hits · 576 pts

Severity	File	Line	Snippet	Context
LOW	build_backend.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	build_web.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_long.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/utils.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_embedding.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_runner.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_latency.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_rerank.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	benchmark/benchmark_serving.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	frontend/next.config.ts	1	// Output mode (build only):	COMMENT
LOW	xinference/isolation.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/conftest.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/fields.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/constants.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/constants.py	161	# XINFERENCE_AUTH_ADVANCED=0/false/no to run with no authentication at all.	COMMENT
LOW	xinference/constants.py	501		COMMENT
LOW	xinference/__init__.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/types.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/utils.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/device_utils.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/_compat.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/ui/__init__.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/worker.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/worker.py	3221		COMMENT
LOW	xinference/core/worker.py	3301	f"[DEBUG] CUDA version check passed: cuda_version={cuda_version}, keeping settings.extra_index_url={sett	COMMENT
LOW	xinference/core/worker.py	3841	# install): allocate_devices/allocate_devices_with_gpu_idx	COMMENT
LOW	xinference/core/worker.py	3861	# afterwards cannot replace them. Spawning before install	COMMENT
LOW	xinference/core/worker.py	5321	await supervisor_ref.call_collective_manager(	COMMENT
LOW	xinference/core/metrics.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/event.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/progress_tracker.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/log.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/monitor_config_store.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/supervisor.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/virtual_env_manager.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/launch_strategy.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/resource.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/__init__.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/cache_tracker.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/autostart.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/http_protocol.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/replica_config.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/model.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/utils.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/status_guard.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/exceptions.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/launch_history_store.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/otel.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/system_settings_store.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_virtual_env_manager.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_model.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_continuous_batching.py	1	# Copyright 2022-2026 XProbe Inc.	COMMENT
LOW	xinference/core/tests/test_utils.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_system_settings_store.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_launch_orphan_cleanup.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_metrics.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_restful_api.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_subpool_hardening.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/test_cache_tracker.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
LOW	xinference/core/tests/__init__.py	1	# Copyright 2022-2026 Xinference Holdings Pte. Ltd	COMMENT
541 more matches not shown…

Deep Nesting649 hits · 542 pts

Severity	File	Line	Context
LOW	build_backend.py	100	CODE
LOW	benchmark/benchmark_runner.py	88	CODE
LOW	benchmark/benchmark_runner.py	206	CODE
LOW	xinference/constants.py	186	CODE
LOW	xinference/utils.py	42	CODE
LOW	xinference/device_utils.py	273	CODE
LOW	xinference/device_utils.py	385	CODE
LOW	xinference/device_utils.py	635	CODE
LOW	xinference/core/worker.py	391	CODE
LOW	xinference/core/worker.py	421	CODE
LOW	xinference/core/worker.py	464	CODE
LOW	xinference/core/worker.py	498	CODE
LOW	xinference/core/worker.py	799	CODE
LOW	xinference/core/worker.py	1108	CODE
LOW	xinference/core/worker.py	1410	CODE
LOW	xinference/core/worker.py	1676	CODE
LOW	xinference/core/worker.py	1737	CODE
LOW	xinference/core/worker.py	1999	CODE
LOW	xinference/core/worker.py	2111	CODE
LOW	xinference/core/worker.py	2275	CODE
LOW	xinference/core/worker.py	2413	CODE
LOW	xinference/core/worker.py	2805	CODE
LOW	xinference/core/worker.py	2954	CODE
LOW	xinference/core/worker.py	3141	CODE
LOW	xinference/core/worker.py	3614	CODE
LOW	xinference/core/worker.py	4296	CODE
LOW	xinference/core/worker.py	4590	CODE
LOW	xinference/core/worker.py	4946	CODE
LOW	xinference/core/worker.py	4976	CODE
LOW	xinference/core/worker.py	5090	CODE
LOW	xinference/core/metrics.py	550	CODE
LOW	xinference/core/metrics.py	1213	CODE
LOW	xinference/core/monitor_config_store.py	156	CODE
LOW	xinference/core/supervisor.py	106	CODE
LOW	xinference/core/supervisor.py	163	CODE
LOW	xinference/core/supervisor.py	599	CODE
LOW	xinference/core/supervisor.py	1371	CODE
LOW	xinference/core/supervisor.py	2269	CODE
LOW	xinference/core/supervisor.py	2424	CODE
LOW	xinference/core/supervisor.py	2594	CODE
LOW	xinference/core/supervisor.py	3196	CODE
LOW	xinference/core/supervisor.py	3521	CODE
LOW	xinference/core/supervisor.py	3625	CODE
LOW	xinference/core/supervisor.py	3877	CODE
LOW	xinference/core/supervisor.py	5659	CODE
LOW	xinference/core/supervisor.py	5743	CODE
LOW	xinference/core/supervisor.py	6106	CODE
LOW	xinference/core/supervisor.py	2906	CODE
LOW	xinference/core/supervisor.py	3247	CODE
LOW	xinference/core/virtual_env_manager.py	405	CODE
LOW	xinference/core/virtual_env_manager.py	1473	CODE
LOW	xinference/core/virtual_env_manager.py	574	CODE
LOW	xinference/core/virtual_env_manager.py	631	CODE
LOW	xinference/core/virtual_env_manager.py	821	CODE
LOW	xinference/core/launch_strategy.py	57	CODE
LOW	xinference/core/router_scheduler.py	214	CODE
LOW	xinference/core/cache_tracker.py	51	CODE
LOW	xinference/core/cache_tracker.py	77	CODE
LOW	xinference/core/cache_tracker.py	109	CODE
LOW	xinference/core/cache_tracker.py	131	CODE
589 more matches not shown…

AI Structural Patterns304 hits · 284 pts

Severity	File	Line	Context
LOW	xinference/core/worker.py	3614	CODE
LOW	xinference/core/worker.py	5235	CODE
LOW	xinference/core/metrics.py	1279	CODE
LOW	xinference/core/supervisor.py	2594	CODE
LOW	xinference/core/supervisor.py	3196	CODE
LOW	xinference/core/supervisor.py	6467	CODE
LOW	xinference/core/supervisor.py	6483	CODE
LOW	xinference/deploy/utils.py	218	CODE
LOW	xinference/model/video/core.py	172	CODE
LOW	xinference/model/llm/memory.py	299	CODE
LOW	xinference/model/llm/lmdeploy/core.py	306	CODE
LOW	xinference/model/llm/vllm/distributed_worker_actor.py	83	CODE
LOW	…nference/model/embedding/sentence_transformers/core.py	339	CODE
LOW	xinference/model/embedding/vllm/core.py	280	CODE
LOW	xinference/model/image/core.py	252	CODE
LOW	xinference/model/image/ocr/paddleocr_vl.py	222	CODE
LOW	xinference/model/image/vllm/core.py	79	CODE
LOW	xinference/model/image/stable_diffusion/core.py	88	CODE
LOW	xinference/model/image/sglang/core.py	64	CODE
LOW	xinference/model/rerank/sentence_transformers/core.py	505	CODE
LOW	xinference/api/restful_api.py	2502	CODE
LOW	xinference/api/restful_api.py	2561	CODE
LOW	xinference/api/restful_api.py	2743	CODE
LOW	xinference/api/restful_api.py	823	CODE
LOW	xinference/api/routers/admin.py	565	CODE
LOW	xinference/api/routers/admin.py	1323	CODE
LOW	xinference/client/restful/restful_client.py	1258	CODE
LOW	xinference/client/restful/async_restful_client.py	1379	CODE
LOW	xinference/thirdparty/cosyvoice/dataset/dataset.py	104	CODE
LOW	xinference/thirdparty/cosyvoice/hifigan/generator.py	397	CODE
LOW	xinference/thirdparty/cosyvoice/transformer/decoder.py	58	CODE
LOW	xinference/thirdparty/cosyvoice/transformer/decoder.py	276	CODE
LOW	…e/thirdparty/cosyvoice/transformer/upsample_encoder.py	107	CODE
LOW	xinference/thirdparty/cosyvoice/transformer/encoder.py	39	CODE
LOW	xinference/thirdparty/cosyvoice/transformer/encoder.py	341	CODE
LOW	xinference/thirdparty/cosyvoice/transformer/encoder.py	390	CODE
LOW	xinference/thirdparty/cosyvoice/utils/mask.py	158	CODE
LOW	xinference/thirdparty/cosyvoice/cli/model.py	170	CODE
LOW	xinference/thirdparty/cosyvoice/cli/model.py	321	CODE
LOW	xinference/thirdparty/cosyvoice/flow/decoder.py	295	CODE
LOW	xinference/thirdparty/cosyvoice/flow/flow.py	25	CODE
LOW	xinference/thirdparty/cosyvoice/flow/flow.py	152	CODE
LOW	xinference/thirdparty/deepseek_vl2/models/siglip_vit.py	194	CODE
LOW	xinference/thirdparty/deepseek_vl2/models/siglip_vit.py	249	CODE
LOW	…party/deepseek_vl2/models/processing_deepseek_vl_v2.py	132	CODE
LOW	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	264	CODE
LOW	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	1472	CODE
LOW	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	1657	CODE
LOW	…ce/thirdparty/deepseek_vl2/models/modeling_deepseek.py	1878	CODE
LOW	…irdparty/deepseek_vl2/models/configuration_deepseek.py	117	CODE
LOW	…rdparty/deepseek_vl2/models/modeling_deepseek_vl_v2.py	134	CODE
LOW	…rdparty/deepseek_vl2/models/modeling_deepseek_vl_v2.py	559	CODE
LOW	…rdparty/deepseek_vl2/models/modeling_deepseek_vl_v2.py	642	CODE
LOW	xinference/thirdparty/audiotools/post.py	90	CODE
LOW	xinference/thirdparty/audiotools/metrics/spectral.py	41	CODE
LOW	xinference/thirdparty/audiotools/metrics/spectral.py	124	CODE
LOW	xinference/thirdparty/audiotools/core/loudness.py	100	CODE
LOW	xinference/thirdparty/audiotools/data/transforms.py	755	CODE
LOW	xinference/thirdparty/audiotools/data/transforms.py	892	CODE
LOW	xinference/thirdparty/audiotools/data/transforms.py	1453	CODE
244 more matches not shown…

Modern Structural Boilerplate253 hits · 256 pts

Severity	File	Line	Snippet	Context
LOW	benchmark/benchmark_long.py	27	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/utils.py	25	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/benchmark_embedding.py	28	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/benchmark_runner.py	28	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/benchmark_latency.py	26	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/benchmark_rerank.py	28	logger = logging.getLogger(__name__)	CODE
LOW	benchmark/benchmark_serving.py	27	logger = logging.getLogger(__name__)	CODE
LOW	xinference/constants.py	20	logger = logging.getLogger(__name__)	CODE
LOW⚡	xinference/device_utils.py	22	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/worker.py	1030	def update_system_settings(self, settings: Dict[str, Any]) -> None:	CODE
LOW⚡	xinference/core/metrics.py	33	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/metrics.py	849	def _update_token_router_gauges(cluster_data: Dict[str, Any]) -> None:	CODE
LOW	xinference/core/metrics.py	1213	def update_security_gauges(auth_service) -> None:	CODE
LOW	xinference/core/router_orchestration.py	22	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/progress_tracker.py	35	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/monitor_config_store.py	22	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/supervisor.py	4849	async def update_system_settings(self, settings: Dict[str, Any]) -> None:	CODE
LOW	xinference/core/virtual_env_manager.py	29	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/http_protocol.py	19	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/model.py	54	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/utils.py	37	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/tokenizer_asset_registry.py	22	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/launch_history_store.py	28	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/otel.py	44	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/system_settings_store.py	30	logger = logging.getLogger(__name__)	CODE
LOW	xinference/core/system_settings_store.py	137	def set_optional(name: str, value: str) -> None:	CODE
LOW	xinference/deploy/worker.py	27	logger = logging.getLogger(__name__)	CODE
LOW	xinference/deploy/supervisor.py	32	logger = logging.getLogger(__name__)	CODE
LOW	xinference/deploy/local.py	36	logger = logging.getLogger(__name__)	CODE
LOW	xinference/deploy/router_agent.py	8	__all__ = ["main"]	CODE
LOW	xinference/deploy/utils.py	45	logger = logging.getLogger(__name__)	CODE
LOW	xinference/deploy/router.py	8	__all__ = ["main"]	CODE
LOW	xinference/model/custom.py	26	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/cache_manager.py	9	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/batch.py	24	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/utils.py	66	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/video/mlx_video.py	42	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/video/core.py	23	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/video/engine_family.py	22	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/video/diffusers.py	49	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/video/tests/test_diffusers_video.py	23	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/reasoning_parser.py	12	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/custom.py	24	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/cache_manager.py	25	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/llm_family.py	43	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/core.py	35	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/utils.py	63	logger = logging.getLogger(__name__)	CODE
LOW	…ence/model/llm/tool_parsers/deepseek_v4_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…rence/model/llm/tool_parsers/minimax_m3_tool_parser.py	10	logger = logging.getLogger(__name__)	CODE
LOW	…ence/model/llm/tool_parsers/deepseek_v3_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/tool_parsers/gemma_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…nference/model/llm/tool_parsers/minimax_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…nference/model/llm/tool_parsers/kimi_k3_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/tool_parsers/glm5_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…ce/model/llm/tool_parsers/deepseek_v3_1_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/tool_parsers/qwen_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…ence/model/llm/tool_parsers/deepseek_r1_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/tool_parsers/glm4_tool_parser.py	8	logger = logging.getLogger(__name__)	CODE
LOW	xinference/model/llm/tool_parsers/llama3_tool_parser.py	9	logger = logging.getLogger(__name__)	CODE
LOW	…ce/model/llm/tool_parsers/deepseek_v3_2_tool_parser.py	8	logger = logging.getLogger(__name__)	CODE
193 more matches not shown…

Dead Code125 hits · 181 pts

Severity	File	Line	Context
MEDIUM	xinference/model/image/ocr/deepdoc.py	338	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	296	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	298	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	299	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	303	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	307	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	309	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	322	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	324	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	326	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	327	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	337	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	342	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	343	CODE
MEDIUM	…ty/sensenova_u1/models/neo_unify/modeling_qwen3_moe.py	344	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	273	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	275	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	276	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	278	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	279	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	281	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	282	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	287	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	288	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	289	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	298	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	300	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	310	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	312	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	313	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	325	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	329	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	353	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	357	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	361	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	362	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	364	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	368	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	369	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	383	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	384	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	385	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	386	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	387	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	388	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	389	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	395	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	396	CODE
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	397	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	775	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	776	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	777	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	779	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	780	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	782	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	784	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	785	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	787	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	788	CODE
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	790	CODE
65 more matches not shown…

Docstring Block Structure36 hits · 180 pts

Severity	File	Line	Snippet	Context
HIGH	xinference/utils.py	87	Recursively convert an object into a hashable form. This function is useful for creating deterministic cache k	STRING
HIGH	xinference/deploy/utils.py	1072	Convert CLI string arguments to appropriate Python types. Handles type conversion for Click command arguments with	STRING
HIGH	…ence/model/llm/tool_parsers/deepseek_v4_tool_parser.py	102	Extract tool calls from complete model output. Args: model_output: The complete output str	STRING
HIGH	…ce/model/llm/tool_parsers/deepseek_v3_2_tool_parser.py	95	Extract tool calls from complete model output. Args: model_output: The complete output str	STRING
HIGH	xinference/model/embedding/core.py	362	Extract the 'sentences' argument and remaining kwargs from (args, *kwargs) for a given function.	STRING
HIGH	xinference/model/rerank/sentence_transformers/core.py	781	Extract the 'documents' and 'query' argument and remaining kwargs from (args, *kwargs) for a given fu	STRING
HIGH	xinference/model/rerank/vllm/core.py	303	Extract the 'documents' and 'query' argument and remaining kwargs from (args, *kwargs) for a given fu	STRING
HIGH	xinference/api/utils.py	101	Get a model with standardized error handling. Replaces the repeated pattern: ```python try: mo	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	18	def subsequent_mask( size: int, device: torch.device = torch.device("cpu"), ) -> torch.Tensor: """C	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	57	Create mask for subsequent steps (size, size). This mask is used only in decoder which works in an auto-regressive	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	95	Create mask for subsequent steps (size, size) with chunk size, this is for streaming encoder Args: s	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	133	Create mask for subsequent steps (size, size) with chunk size, this is for streaming encoder Args: s	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/mask.py	240	Make mask tensor containing indices of padded part. See description of make_non_pad_mask. Args: length	STRING
HIGH	xinference/thirdparty/cosyvoice/utils/common.py	30	Perform padding for the list of tensors. Args: xs (List): List of Tensors [(T_1, ``), (T_2, ``), ..., (T_	STRING
HIGH	xinference/thirdparty/cosyvoice/flow/decoder.py	211	Forward pass of the UNet1DConditional model. Args: x (torch.Tensor): shape (batch_size, in_channels	STRING
HIGH	xinference/thirdparty/cosyvoice/flow/decoder.py	406	Forward pass of the UNet1DConditional model. Args: x (torch.Tensor): shape (batch_size, in_channels	STRING
HIGH	…u1/prompt_enhance/adapters/chat_completions_adapter.py	154	Extract the assistant message text from a chat/completions response. Handles both plain-string and list-of-cont	STRING
HIGH	…u1/prompt_enhance/adapters/chat_completions_adapter.py	191	Call the ``/chat/completions`` endpoint with vision content. Args: user_prompt: User-facing text in	STRING
HIGH	…irdparty/sensenova_u1/prompt_enhance/adapters/utils.py	13	Read raw image bytes from a path or return bytes unchanged. Args: image: File path to an image, or raw imag	STRING
HIGH	…senova_u1/prompt_enhance/adapters/anthropic_adapter.py	160	Extract the assistant message text from Anthropic Messages response. Handles responses with or without content	STRING
HIGH	…senova_u1/prompt_enhance/adapters/anthropic_adapter.py	193	Call the ``/v1/messages`` endpoint with vision content. Args: user_prompt: User-facing text instruc	STRING
HIGH	…ce/thirdparty/ace_step_1_5/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	…/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	…1_5/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	27	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	…tep_1_5/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	…e_step_1_5/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	…_1_5/acestep/core/generation/handler/generate_music.py	252	Generate audio from text/reference inputs and return response payload. Args: captions: Text prompt	STRING
HIGH	…party/megatts3/tts/modules/ar_dur/commons/seq_utils.py	113	Make mask tensor containing indices of padded part. Args: lengths (LongTensor or List): Batch of lengths (B,	STRING
HIGH	…party/megatts3/tts/modules/ar_dur/commons/seq_utils.py	219	Make mask tensor containing indices of non-padded part. Args: lengths (LongTensor or List): Batch of lengths	STRING
HIGH	…ference/thirdparty/matcha/models/components/decoder.py	364	Forward pass of the UNet1DConditional model. Args: x (torch.Tensor): shape (batch_size, in_channels	STRING
HIGH	xinference/thirdparty/indextts/utils/common.py	86	Make mask tensor containing indices of padded part. See description of make_non_pad_mask. Args: length	STRING
HIGH	xinference/thirdparty/indextts/utils/utils.py	54	Make mask tensor containing indices of padded part. See description of make_non_pad_mask. Args: length	STRING

Self-Referential Comments47 hits · 142 pts

Severity	File	Line	Snippet	Context
MEDIUM	xinference/core/virtual_env_manager.py	813	# This method is deprecated and no longer needed	COMMENT
MEDIUM⚡	xinference/core/tests/test_restful_api.py	1645	# Create a mock handler with the conversion method	COMMENT
MEDIUM⚡	xinference/core/tests/test_restful_api.py	1652	# Create a temporary RESTfulAPI instance to access the method	COMMENT
MEDIUM⚡	xinference/deploy/test/test_log_rotation.py	211	# Create a file dated 100 days ago	COMMENT
MEDIUM	xinference/model/utils.py	3068	# Create a mapping of updated_at to model for comparison	COMMENT
MEDIUM	xinference/model/utils.py	3143	# Create a copy of built-in models for merging	COMMENT
MEDIUM	xinference/model/llm/__init__.py	297	# Create a copy of built-in models for merging	COMMENT
MEDIUM	xinference/model/llm/vllm/distributed_executor_v1.py	101	# Create the parallel GPU workers.	COMMENT
MEDIUM	xinference/model/llm/vllm/distributed_executor.py	146	# Create the parallel GPU workers.	COMMENT
MEDIUM	xinference/model/llm/transformers/chatglm.py	126	# This method is adapted from https://github.com/THUDM/GLM-4/blob/main/basic_demo/openai_api_server.py	COMMENT
MEDIUM⚡	xinference/model/llm/transformers/core.py	64	# Define the decorator to support multiple names registration	COMMENT
MEDIUM⚡	…erence/model/llm/transformers/tests/test_tensorizer.py	100	# Create a dummy cache file to simulate cache existence	COMMENT
MEDIUM	…ence/model/embedding/vllm/tests/test_vllm_embedding.py	104	# Create a long text that exceeds 512 tokens	COMMENT
MEDIUM	xinference/model/audio/utils.py	178	# Create the file, close our handle immediately, and clean it up	COMMENT
MEDIUM	xinference/model/rerank/sentence_transformers/core.py	311	# Create a simple wrapper for return value conversion only.	COMMENT
MEDIUM	xinference/thirdparty/audiotools/ml/accelerator.py	14	"""This class is used to prepare models and dataloaders for	STRING
MEDIUM	xinference/thirdparty/fish_speech/tools/run_webui.py	73	# Create the inference engine	COMMENT
MEDIUM	…rence/thirdparty/fish_speech/tools/vqgan/extract_vq.py	24	# This file is used to convert the audio files to text files using the Whisper model.	COMMENT
MEDIUM	…fish_speech/tools/server/agent/pre_generation_utils.py	39	# Create a conversation object and encode it for inference	COMMENT
MEDIUM	…fish_speech/tools/inference_engine/reference_loader.py	29	# Define the torchaudio backend	COMMENT
MEDIUM	…rdparty/fish_speech/tools/inference_engine/__init__.py	166	# Create a queue to get the response	COMMENT
MEDIUM	…erence/thirdparty/fish_speech/fish_speech/i18n/scan.py	55	# Define the standard file name	COMMENT
MEDIUM	…erence/thirdparty/fish_speech/fish_speech/i18n/scan.py	78	# Define the standard file name	COMMENT
MEDIUM	…arty/sensenova_u1/models/neo_unify/modeling_neo_vit.py	38	# Create the batch index for each patch (B x patch count)	COMMENT
MEDIUM	…rty/sensenova_u1/models/neo_unify/modeling_neo_chat.py	134	# Create the batch index for each patch (B x patch count)	COMMENT
MEDIUM	…dparty/sensenova_u1/models/neo_unify/modeling_qwen3.py	1147	# Create the masks	COMMENT
MEDIUM	xinference/thirdparty/melo/text/japanese.py	540	# Initialize kakasi object	COMMENT
MEDIUM	xinference/thirdparty/melo/text/japanese.py	636	# # Initialize kakasi object	COMMENT
MEDIUM	xinference/thirdparty/melo/text/korean.py	181	# # Initialize kakasi object	COMMENT
MEDIUM	…ference/thirdparty/ace_step_1_5/acestep/audio_utils.py	83	# Create a copy to avoid modifying original in-place	COMMENT
MEDIUM	xinference/thirdparty/ace_step_1_5/acestep/inference.py	960	# Create a copy of params dict for this audio	COMMENT
MEDIUM	xinference/thirdparty/f5_tts/infer/utils_infer.py	509	# Create a combined spectrogram	COMMENT
MEDIUM	xinference/thirdparty/deepseek_vl/serve/app_deepseek.py	96	# Create a copy of the conversation to avoid history truncation in the UI	COMMENT
MEDIUM⚡	xinference/thirdparty/megatts3/tts/infer_cli.py	49	# Define the output path with a .wav extension	COMMENT
MEDIUM⚡	…erence/thirdparty/megatts3/tts/utils/audio_utils/io.py	63	# Define the output path with a .wav extension	COMMENT
MEDIUM	xinference/thirdparty/mlx/flux/flux.py	142	# Create the latent variables	COMMENT
MEDIUM	xinference/thirdparty/matcha/onnx/export.py	164	# Create the output directory (if not exists)	COMMENT
MEDIUM	xinference/thirdparty/indextts/infer_v2.py	1134	# Create a temporary file for the complete audio	COMMENT
MEDIUM	…irdparty/indextts/gpt/transformers_generation_utils.py	431	# Create the causal mask with fixed shape in advance, to reduce recompilations. If the function to create	COMMENT
MEDIUM	xinference/thirdparty/indextts/BigVGAN/ECAPA_TDNN.py	246	"""This class implements an attentive statistic pooling layer for each channel.	STRING
MEDIUM	xinference/thirdparty/indextts/BigVGAN/ECAPA_TDNN.py	585	"""This class implements the cosine similarity on the top of features.	STRING
MEDIUM	xinference/thirdparty/indextts/BigVGAN/nnet/CNN.py	24	"""This function implements SincConv (SincNet).	STRING
MEDIUM	xinference/thirdparty/indextts/BigVGAN/nnet/CNN.py	306	"""This function implements 1d convolution.	STRING
MEDIUM	…/utils/maskgct/models/codec/facodec/facodec_trainer.py	521	# Create the normalized F0 sequence with unvoiced frames	COMMENT
MEDIUM	xinference/thirdparty/indextts/s2mel/modules/commons.py	382	# Create the normalized F0 sequence with unvoiced frames	COMMENT
MEDIUM	doc/source/gen_docs.py	67	# Create a proper ModuleSpec that importlib.util.find_spec can find	COMMENT
MEDIUM	doc/source/norm_zh.py	15	# This file is derived from https://github.com/mars-project/mars/blob/master/docs/source/norm_zh.py	COMMENT

Cross-Language Confusion25 hits · 141 pts

Severity	File	Line	Snippet	Context
HIGH	benchmark/benchmark_runner.py	142	# "finish_reason": null,	COMMENT
HIGH	xinference/deploy/docker/pypiserver/selfcheck.py	80	# live in a writable directory (not /dev/null).	COMMENT
HIGH⚡	xinference/deploy/test/test_utils.py	218	arg = '{"int_val": 10, "float_val": 1.5, "bool_val": false, "null_val": null}'	CODE
HIGH	xinference/deploy/test/test_utils.py	324	result = handle_click_args_type('[1, "two", null, true, {"key": "value"}]')	CODE
HIGH	xinference/model/llm/tool_parsers/llama3_tool_parser.py	53	# (e.g., True/False/None instead of true/false/null).	COMMENT
HIGH⚡	…odel/llm/tool_parsers/tests/test_llama3_tool_parser.py	41	model_output = '{"name": "test_func", "parameters": {"value": null}}'	CODE
HIGH	xinference/model/llm/tests/test_utils.py	2614	"arguments": None, # null, left as-is	CODE
HIGH	xinference/model/llm/tests/test_utils.py	2632	"arguments": "null", # valid JSON null, left as-is	CODE
HIGH⚡	xinference/model/image/ocr/tests/test_deepdoc_parse.py	167	# The field is omitted rather than set to null, so clients can rely	COMMENT
HIGH	xinference/model/audio/minimax_music3.py	170	"MiniMax-Music3 only accepts `voice` as null, an empty string, "	CODE
HIGH	xinference/model/audio/ace_step.py	349	"ACE-Step 1.5 only accepts `voice` as null, an empty string, "	CODE
HIGH	xinference/api/restful_api.py	955	"npm ci && npm run build" (this stages the export at the directory	CODE
HIGH	…nference/api/tests/test_frontend_static_real_export.py	18	(``cd frontend && npm run build``) and runs this module against	STRING
HIGH	xinference/api/tests/test_anthropic_messages.py	145	'"finish_reason":null}]}\n\n'	CODE
HIGH	xinference/api/tests/test_token_router_dispatch.py	1272	yield b'data: {"choices":[{"delta":{"content":"hello"},"finish_reason":null}]}\n\n'	CODE
HIGH	xinference/thirdparty/audiotools/preference.py	179	if (region != null) {	CODE
HIGH	…ference/thirdparty/fish_speech/tools/webui/__init__.py	18	js="() => {const params = new URLSearchParams(window.location.search);if (!params.has('__theme')) {params.se	CODE
HIGH	…rence/thirdparty/ace_step_1_5/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	…rence/thirdparty/ace_step_1_5/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	…ce_step_1_5/acestep/models/common/flow_edit_helpers.py	50	"""Reduce a ``(cond, null)``-packed prediction to a guided velocity.	STRING
HIGH	…ce_step_1_5/acestep/models/common/flow_edit_helpers.py	184	torch.cat([enc_hs, null], dim=0),	CODE
HIGH	…odels/codec/speechtokenizer/modules/quantization/ac.py	138	self.packer.push(b1)	CODE
HIGH	…odels/codec/speechtokenizer/modules/quantization/ac.py	185	self.packer.push(b1)	CODE
HIGH	…odels/codec/speechtokenizer/modules/quantization/ac.py	304	encoder.push(symbol, q_cdf)	CODE
HIGH	…nce/thirdparty/indextts/s2mel/modules/flow_matching.py	88	# Stack original and CFG (null) inputs for batched processing	COMMENT

Redundant / Tautological Comments95 hits · 128 pts

Severity	File	Line	Snippet	Context
LOW	xinference/core/worker.py	3223	# Check if packages contain PyTorch system markers (#system_torch#, etc.)	COMMENT
LOW	xinference/core/worker.py	3784	# Check if cancelled while waiting in queue	COMMENT
LOW	xinference/model/utils.py	2909	# Check if flash_attn is installed	COMMENT
LOW	xinference/model/utils.py	2917	# Check if CUDA is available	COMMENT
LOW	xinference/model/llm/reasoning_parser.py	55	# Check if <think> is present in previous or delta.	COMMENT
LOW	xinference/model/llm/llm_family.py	407	# Append the class to the list of transformer classes	COMMENT
LOW	xinference/model/llm/harmony.py	60	# Check if buffer ends with partial 'assistantfinal'	COMMENT
LOW	xinference/model/llm/harmony.py	76	# Check if this is actually a new message starting with 'analysis'	COMMENT
LOW	xinference/model/llm/harmony.py	102	# Check if buffer ends with partial 'assistantfinal'	COMMENT
LOW	xinference/model/llm/tool_parsers/qwen_tool_parser.py	360	# Check if current output contains tool_call start token	COMMENT
LOW	…ence/model/llm/tool_parsers/deepseek_r1_tool_parser.py	86	# Check if this block is a tool call	COMMENT
LOW	xinference/model/llm/tests/test_llm_model.py	311	# Check if tool_calls is present and is a list	COMMENT
LOW	xinference/model/llm/tests/test_llm_model.py	325	# Check if arguments is a valid JSON string	COMMENT
LOW	xinference/model/llm/vllm/xavier/engine.py	125	# Check if we have a cached last_output from the previous iteration.	COMMENT
LOW	…ce/model/llm/vllm/patches/hybrid_kv_cache_page_size.py	107	# Check if the original problematic code exists	COMMENT
LOW	…ce/model/llm/vllm/patches/hybrid_kv_cache_page_size.py	119	# Check if all smaller pages divide max evenly	COMMENT
LOW	xinference/model/llm/transformers/core.py	469	# Check if streaming	COMMENT
LOW	xinference/model/llm/transformers/core.py	1060	# Check if streaming	COMMENT
LOW⚡	…erence/model/llm/transformers/tests/test_tensorizer.py	118	# Check if the logger.info was called with the expected message, indicating early return due to cache existence	COMMENT
LOW	xinference/model/llm/mlx/core.py	179	# Check if next_generated method exists (new API)	COMMENT
LOW	xinference/model/llm/mlx/core.py	440	# Check if there are any pending results (early results that arrived before queue was ready)	COMMENT
LOW	xinference/model/llm/mlx/core.py	492	# Check if generation is finished	COMMENT
LOW	xinference/model/llm/mlx/core.py	517	# Check if request is still in active set	COMMENT
LOW	xinference/model/llm/mlx/core.py	2102	# Check if processor supports audio to avoid feature_extractor errors	COMMENT
LOW	xinference/model/embedding/core.py	216	# Check if sentences is a two-dimensional list of integers	COMMENT
LOW	xinference/model/image/stable_diffusion/core.py	916	# Check if we have a cancel callback for this request	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/run_webui.py	48	# Check if MPS or CUDA is available	COMMENT
LOW	…e/thirdparty/fish_speech/tools/server/model_manager.py	37	# Check if MPS or CUDA is available	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/server/views.py	159	# Check if the text is too long	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/server/views.py	166	# Check if streaming is enabled	COMMENT
LOW	…/thirdparty/fish_speech/tools/server/agent/generate.py	64	# Check if all samples are finished	COMMENT
LOW	…rty/fish_speech/tools/server/agent/generation_utils.py	102	# Check if the token is semantic	COMMENT
LOW	…nference/thirdparty/ace_step_1_5/acestep/gpu_config.py	1217	# Check if a per-process memory fraction has been set	COMMENT
LOW	…nference/thirdparty/ace_step_1_5/acestep/gpu_config.py	1400	# Check if model size is in available models	COMMENT
LOW	…rence/thirdparty/ace_step_1_5/acestep/llm_inference.py	1688	# Check if user provided a meaningful negative prompt (not the default)	COMMENT
LOW	…rence/thirdparty/ace_step_1_5/acestep/llm_inference.py	2889	# Check if this is a new field (no leading spaces and contains ':')	COMMENT
LOW	…rence/thirdparty/ace_step_1_5/acestep/llm_inference.py	2979	# Check if keys need "model." prefix by inspecting first key	COMMENT
LOW	…rence/thirdparty/ace_step_1_5/acestep/llm_inference.py	3287	# Check if all items are done	COMMENT
LOW	…ce/thirdparty/ace_step_1_5/acestep/model_downloader.py	646	# Check if this is a known LM model	COMMENT
LOW	…ce/thirdparty/ace_step_1_5/acestep/model_downloader.py	648	# Check if it might be a variant name	COMMENT
LOW	…ce/thirdparty/ace_step_1_5/acestep/model_downloader.py	689	# Check if this is the default turbo model (part of main)	COMMENT
LOW	…ce/thirdparty/ace_step_1_5/acestep/model_downloader.py	693	# Check if it's a known sub-model	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	742	# Check if LM generation failed	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	913	# Check if generation failed	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1154	# Check if LLM is initialized	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1178	# Check if LLM returned empty metadata (error case)	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1331	# Check if LLM is initialized	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1353	# Check if LLM returned empty metadata (error case)	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1505	# Check if LLM is initialized	COMMENT
LOW	xinference/thirdparty/ace_step_1_5/acestep/inference.py	1527	# Check if LLM returned empty metadata (error case)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	751	# Check if first token starts with a note (A-G, case insensitive, with optional leading space)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1804	# Check if field is user-provided and we haven't started injecting yet	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1260	# Check if new_prefix is a valid prefix in the active trie	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1389	# Check if decoded token matches the prefix start (allowing for space prefixes)	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1424	# Check if this digit could lead to a valid final value	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1433	# Check if we can still reach min_val	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1597	# Check if input contains </think> token sequence	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1750	# Check if we have tokens in queue for user-provided field	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1763	# Check if we should stop at reasoning (after </think> tag)	COMMENT
LOW	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1766	# Check if the next token would complete the fixed string	COMMENT
35 more matches not shown…

Modern AI Meta-Vocabulary15 hits · 52 pts

Severity	File	Line	Snippet	Context
MEDIUM	xinference/core/tests/test_model.py	276	# rerank/embedding model must run del + empty_cache so its VRAM is	COMMENT
MEDIUM	xinference/core/tests/test_worker.py	2074	# test terminate LLM model, then launch embedding model	COMMENT
MEDIUM⚡	xinference/model/embedding/embed_family.py	115	# { embedding model name -> { engine name -> engine params } }	COMMENT
MEDIUM	xinference/model/embedding/core.py	90	# this class define the basic info of embedding model	COMMENT
MEDIUM	…nference/model/embedding/sentence_transformers/core.py	528	# seems already support prompt in embedding model	COMMENT
MEDIUM⚡	xinference/api/restful_api.py	845	# Attach API instance for dependency injection (Depends(get_api), etc.)	COMMENT
MEDIUM	xinference/api/schemas/requests.py	33	# <0 = cap at the model's max_tokens. Honored by the embedding model base	COMMENT
MEDIUM	…ference/thirdparty/deepseek_vl2/models/conversation.py	33	# The number of few shot examples	COMMENT
MEDIUM	…irdparty/sensenova_u1/models/neo_unify/conversation.py	52	# The number of few shot examples	COMMENT
MEDIUM	xinference/thirdparty/deepseek_vl/utils/conversation.py	65	# The number of few shot examples	COMMENT
MEDIUM	xinference/thirdparty/internvl/conversation.py	50	# The number of few shot examples	COMMENT
MEDIUM	xinference/thirdparty/whisper/transcribe.py	402	# skip silence before possible hallucinations	COMMENT
MEDIUM⚡	xinference/thirdparty/whisper/transcribe.py	414	# if first segment might be a hallucination, skip leading silence	COMMENT
MEDIUM⚡	xinference/thirdparty/whisper/transcribe.py	422	# skip silence before any possible hallucination that is surrounded	COMMENT
MEDIUM⚡	xinference/thirdparty/whisper/transcribe.py	423	# by silence or more hallucinations	COMMENT

AI Slop Vocabulary22 hits · 45 pts

Severity	File	Line	Snippet	Context
LOW	xinference/core/supervisor.py	3403	# Not launched perhaps, just return 0.0 to prevent error	COMMENT
LOW	xinference/model/utils.py	3062	# New model from user, just add it	COMMENT
LOW⚡	xinference/model/llm/llm_family.py	633	# TODO: If user does not specify quantization, just use the first one	COMMENT
MEDIUM	xinference/model/llm/core.py	252	# Use model_specs from huggingface, as HuggingFace is the most comprehensive.	COMMENT
LOW⚡	xinference/model/embedding/embed_family.py	105	# TODO: If user does not specify quantization, just use the first one	COMMENT
LOW	xinference/model/rerank/rerank_family.py	104	# TODO: If user does not specify quantization, just use the first one	COMMENT
LOW	…erence/thirdparty/cosyvoice/transformer/convolution.py	124	# It's better we just return None if no cache is required,	COMMENT
MEDIUM	xinference/thirdparty/cosyvoice/utils/executor.py	45	# A context manager to be used in conjunction with an instance of	COMMENT
MEDIUM	xinference/thirdparty/cosyvoice/utils/executor.py	97	# A context manager to be used in conjunction with an instance of	COMMENT
MEDIUM	xinference/thirdparty/sensenova_u1/utils/param_count.py	166	# parameter, which is robust to mixed-dtype checkpoints (e.g. norms	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1793	# Avoid infinite recursion: if we're still in a fixed_strings state, just return scores	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	1795	# This shouldn't happen, but if it does, just return scores to avoid recursion	COMMENT
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	…irdparty/ace_step_1_5/acestep/core/scoring/lm_score.py	339	# Calculate recall using the robust logic	COMMENT
LOW	xinference/thirdparty/f5_tts/model/cfm.py	136	duration = torch.maximum(lens + 1, duration) # just add one token so something is generated	CODE
LOW	xinference/thirdparty/f5_tts/model/dataset.py	214	# if want to have different batches between epochs, may just set a seed and log it in ckpt	COMMENT
MEDIUM	xinference/thirdparty/f5_tts/infer/speech_edit.py	54	# [leverage https://github.com/MahmoudAshraf97/ctc-forced-aligner to get char level alignment]	COMMENT
MEDIUM	…irdparty/indextts/gpt/transformers_generation_utils.py	372	# (this alternative is not as robust as calling `generate` and letting it create `cache_position`)	COMMENT
MEDIUM	…irdparty/indextts/gpt/transformers_generation_utils.py	648	# 1. Check whether the user has defined `decoder_input_ids` manually. To facilitate in terms of input naming,	COMMENT
LOW	…ce/thirdparty/indextts/gpt/transformers_beam_search.py	854	# generation. In these cases we simply return the highest scoring outputs.	STRING
LOW	…thirdparty/indextts/gpt/transformers_modeling_utils.py	5402	# We can probably just use the multi-head attention module of PyTorch >=1.1.0	STRING
LOW	xinference/thirdparty/indextts/gpt/conformer_encoder.py	146	# It's better we just return None if no cache is required,	COMMENT

Structural Annotation Overuse28 hits · 43 pts

Severity	File	Line	Snippet	Context
LOW⚡	…m/tool_parsers/tests/test_deepseek_v3_1_tool_parser.py	281	# Step 1: Part 1 of end token	COMMENT
LOW⚡	…m/tool_parsers/tests/test_deepseek_v3_1_tool_parser.py	288	# Step 2: Part 2 of end token	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1015	# Step 1: Get the initial response with stream=True	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1033	# Step 2: Get the final response after tool calls with stream=True	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1051	# Step 3: Get the initial response with stream=False	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1068	# Step 4: Get the final response after tool calls with stream=False	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	106	# Step 1: Encode audio using VQGAN	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	135	# Step 2: Prepare LLM request	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	182	# Step 3: Stream LLM response and decode audio	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	719	# Step 1: Encode full string "keyscale: {keyscale}" (with space, as tokenizer sees it)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	723	# Step 2: Find where context ends in full_token_ids	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	738	# Step 3: Extract keyscale value tokens (everything after context)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	741	# Step 4: Verify we extracted some tokens (sanity check)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	747	# Step 5: Verify first token is a note (A-G)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	759	# Step 6: Build prefix mappings from keyscale value tokens	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	329	# Step 1: Self-attention with AdaLN	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	339	# Step 2: Cross-attention	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	350	# Step 3: MLP with AdaLN	COMMENT
LOW	…irdparty/indextts/gpt/transformers_generation_utils.py	3549	# IMPORTANT: Note that this should appear BEFORE the call to _reorder_cache() to save the maximum memory	STRING
LOW	…irdparty/indextts/gpt/transformers_generation_utils.py	3842	# IMPORTANT: Note that this should appear BEFORE the call to _reorder_cache() to save the maximum memory	STRING
LOW	…irdparty/indextts/gpt/transformers_generation_utils.py	4080	# IMPORTANT: Note that this should appear BEFORE the call to _reorder_cache() to save the maximum memory	STRING

Verbosity Indicators25 hits · 42 pts

Severity	File	Line	Snippet	Context
LOW⚡	…m/tool_parsers/tests/test_deepseek_v3_1_tool_parser.py	281	# Step 1: Part 1 of end token	COMMENT
LOW⚡	…m/tool_parsers/tests/test_deepseek_v3_1_tool_parser.py	288	# Step 2: Part 2 of end token	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1015	# Step 1: Get the initial response with stream=True	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1033	# Step 2: Get the final response after tool calls with stream=True	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1051	# Step 3: Get the initial response with stream=False	COMMENT
LOW	xinference/model/llm/tests/test_stream_options.py	1068	# Step 4: Get the final response after tool calls with stream=False	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	106	# Step 1: Encode audio using VQGAN	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	135	# Step 2: Prepare LLM request	COMMENT
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	182	# Step 3: Stream LLM response and decode audio	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	719	# Step 1: Encode full string "keyscale: {keyscale}" (with space, as tokenizer sees it)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	723	# Step 2: Find where context ends in full_token_ids	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	738	# Step 3: Extract keyscale value tokens (everything after context)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	741	# Step 4: Verify we extracted some tokens (sanity check)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	747	# Step 5: Verify first token is a note (A-G)	COMMENT
LOW⚡	…y/ace_step_1_5/acestep/constrained_logits_processor.py	759	# Step 6: Build prefix mappings from keyscale value tokens	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	156	# Step 1: Concatenate hidden states and masks along sequence dimension	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	162	# Step 2: Sort indices so that mask values of 1 come before 0	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	165	# Step 3: Reorder hidden states using sorted indices	COMMENT
LOW⚡	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	168	# Step 4: Create new mask based on valid sequence lengths	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	500	# Step 1: Self-attention with adaptive layer norm (AdaLN)	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	516	# Step 2: Cross-attention (if enabled) for conditioning on encoder outputs	COMMENT
LOW	…1_5/acestep/models/turbo/modeling_acestep_v15_turbo.py	531	# Step 3: Feed-forward (MLP) with adaptive layer norm	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	329	# Step 1: Self-attention with AdaLN	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	339	# Step 2: Cross-attention	COMMENT
LOW⚡	…hirdparty/ace_step_1_5/acestep/models/mlx/dit_model.py	350	# Step 3: MLP with AdaLN	COMMENT

Synthetic Comment Markers1 hit · 11 pts

Severity	File	Line	Snippet	Context
HIGH⚡	xinference/api/tests/test_ocr_pdf.py	531	# 200 A4 pages at zoomin 3 is only 0.902 G px as requested, so the	COMMENT

Slop Phrases4 hits · 10 pts

Severity	File	Line	Snippet	Context
MEDIUM	xinference/thirdparty/melo/text/chinese.py	73	assert len(word2ph) == len(text) # Sometimes it will crash,you can add a try-catch.	CODE
MEDIUM	xinference/thirdparty/melo/text/chinese_mix.py	80	# assert len(word2ph) == len(text) # Sometimes it will crash,you can add a try-catch.	COMMENT
LOW⚡	…rence/thirdparty/ace_step_1_5/acestep/llm_inference.py	2907	# Don't forget to save the last field	COMMENT
MEDIUM	…nce/thirdparty/f5_tts/train/datasets/prepare_emilia.py	2	# if use updated new version, i.e. WebDataset, feel free to modify / draft your own script	COMMENT

Hallucination Indicators1 hit · 10 pts

Severity	File	Line	Snippet	Context
CRITICAL	xinference/thirdparty/audiotools/core/util.py	475	batch[k] = torch.utils.data._utils.collate.default_collate(v)	CODE

Magic Placeholder Names1 hit · 5 pts

Severity	File	Line	Snippet	Context
HIGH	xinference/thirdparty/fish_speech/tools/api_client.py	176	"authorization": "Bearer YOUR_API_KEY",	CODE

Fake / Example Data3 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW⚡	doc/source/norm_zh.py	62	... "Lorem ipsum dolor sit amet, consectetur adipisicing elit, "	CODE
LOW⚡	doc/source/norm_zh.py	62	... "Lorem ipsum dolor sit amet, consectetur adipisicing elit, "	CODE
LOW⚡	doc/source/norm_zh.py	66	" \"Lorem ipsum dolor sit "	CODE

Example Usage Blocks2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	xinference/thirdparty/fish_speech/tools/fish_e2e.py	263	# Example usage:	COMMENT
LOW	…party/matcha/utils/get_durations_from_trained_model.py	192	# Example usage:	STRING

Overly Generic Function Names1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	…ce/thirdparty/fish_speech/tools/llama/build_dataset.py	65	def run_task(task):	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings