InternLM/lmdeploy

16.8

Adjusted Score

16.8

Raw Score

100%

Time Factor

2026-08-28

Last Push

8.0K

Stars

Python

Language

313.9K

Lines of Code

1.6K

Files

4.1K

Pattern Hits

2026-08-29

Scan Date

0.08

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 0HIGH 125MEDIUM 348LOW 3633

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 4106 distinct pattern matches across 21 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 Identifiers2179 hits · 2073 pts

Severity	File	Line	Snippet	Context
LOW	benchmark/benchmark_chat_completion.py	175	def _dataset_longest_prefix_match(dataset: str, selected: Sequence[str]) -> str \| None:	CODE
LOW	benchmark/benchmark_chat_completion.py	360	def _cached_tokens_from_usage(usage: dict[str, Any] \| None) -> int:	CODE
LOW	benchmark/profile_restful_api.py	159	async def async_request_openai_completions(	CODE
LOW	benchmark/profile_restful_api.py	237	async def async_request_openai_chat_completions(	CODE
LOW	benchmark/profile_restful_api.py	345	async def async_request_sglang_generate(	CODE
LOW	benchmark/profile_restful_api.py	847	def _gen_random_image_data_uri(width: int = width, height: int = height) -> tuple[Image.Image, str, int]:	CODE
LOW⚡	lmdeploy/deepseek_v4_encoding.py	114	def tool_calls_from_openai_format(tool_calls):	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	125	def tool_calls_to_openai_format(tool_calls):	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	469	def sort_tool_results_by_call_order(messages: list[dict[str, Any]]) -> list[dict[str, Any]]:	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	694	def parse_message_from_completion_text(text: str, thinking_mode: str) -> dict[str, Any]:	CODE
LOW	lmdeploy/tokenizer.py	70	def _check_transformers_version(self, model_dir: str, trust_remote_code: bool = False):	CODE
LOW	lmdeploy/tokenizer.py	234	def _convert_tokens_to_string_with_added_encoders(	CODE
LOW	lmdeploy/utils.py	638	def init_custom_process_group(backend=None,	CODE
LOW	lmdeploy/messages.py	197	def convert_stop_bad_words_to_ids(self, tokenizer: Tokenizer):	CODE
LOW⚡	lmdeploy/deepseek_v32_encoding.py	73	def tool_calls_from_openai_format(tool_calls):	CODE
LOW⚡	lmdeploy/deepseek_v32_encoding.py	82	def tool_calls_to_openai_format(tool_calls):	CODE
LOW	lmdeploy/deepseek_v32_encoding.py	357	def parse_message_from_completion_text(text: str, thinking_mode: str):	CODE
LOW	lmdeploy/metrics/metrics_processor.py	101	def increase_succeeded_requests(self):	CODE
LOW	lmdeploy/metrics/metrics_processor.py	114	def increase_api_routed_requests(self):	CODE
LOW	lmdeploy/metrics/metrics_processor.py	118	def decrease_api_routed_requests(self):	CODE
LOW	lmdeploy/metrics/stats.py	90	def update_from_schedule_metrics(self, scheduled_metrics: ScheduleMetrics):	CODE
LOW	lmdeploy/lite/apis/calibrate.py	184	def make_compatible_internvl_config(model_path):	CODE
LOW	lmdeploy/lite/apis/calibrate.py	194	def _get_non_default_generation_parameters(self):	CODE
LOW	lmdeploy/lite/utils/calib_dataloader.py	250	def get_neuralmagic_calibration(dataset, tokenizer, nsamples, seed, seqlen):	CODE
LOW	lmdeploy/lite/utils/memory_efficient.py	45	def find_modules_by_return_value(model: nn.Module, value: str) -> list[nn.Module]:	CODE
LOW	lmdeploy/lite/utils/memory_efficient.py	197	def memory_efficient_inference(model: nn.Module, offload: bool = True, device: str = 'cuda') -> None:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	20	def cal_qparams_per_channel_absmax(w: torch.Tensor, n_bits: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	36	def cal_qparams_per_channel_minmax(w: torch.Tensor, n_bits: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	58	def cal_qparams_per_group_absmax(w: torch.Tensor, n_bits: int, group_size: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	79	def cal_qparams_per_group_minmax(w: torch.Tensor, n_bits: int, group_size: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	105	def cal_qparams_per_tensor_minmax(w: torch.Tensor, n_bits: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/cal_qparams.py	125	def cal_qparams_per_tensor_absmax(w: torch.Tensor, n_bits: int, return_stats: bool = False) -> QParams:	CODE
LOW	lmdeploy/lite/utils/batch_split.py	7	def split_decoder_layer_inputs(batch_size, *args: torch.Tensor \| Any,	CODE
LOW	lmdeploy/lite/utils/batch_split.py	64	def concat_decoder_layer_outputs(batch_outputs: list[Any]) -> Any:	CODE
LOW	lmdeploy/lite/quantization/calibration.py	407	def _wrap_decoder_layers_for_search(self):	CODE
LOW	lmdeploy/cli/utils.py	503	def allow_terminate_by_client(parser):	CODE
LOW	lmdeploy/cli/utils.py	590	def prefix_cache_state_budget(parser):	CODE
LOW	lmdeploy/cli/utils.py	601	def prefix_cache_decode_state_interval(parser):	CODE
LOW	lmdeploy/cli/utils.py	645	def cudagraph_capture_batch_sizes(parser):	CODE
LOW	lmdeploy/cli/utils.py	783	def dllm_confidence_threshold(parser):	CODE
LOW	lmdeploy/cli/utils.py	791	def enable_return_routed_experts(parser):	CODE
LOW	lmdeploy/cli/utils.py	821	def distributed_executor_backend(parser):	CODE
LOW	lmdeploy/vl/engine.py	15	def _get_hf_config_mm_feature_dtype(hf_config) -> torch.dtype \| None:	CODE
LOW	lmdeploy/vl/engine.py	44	def _resolve_mm_feature_dtype(hf_config, backend_config) -> torch.dtype \| None:	CODE
LOW	lmdeploy/vl/engine.py	63	def _raise_exception_on_finish(task: asyncio.Task) -> None:	CODE
LOW	lmdeploy/vl/model/internvl.py	21	def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):	CODE
LOW	lmdeploy/vl/model/preprocess_utils.py	77	def _expand_bundled_image_items(item: dict, token_id: int) -> list[dict]:	CODE
LOW	lmdeploy/vl/model/preprocess_utils.py	107	def _expand_bundled_video_items(item: dict, token_id: int) -> list[dict]:	CODE
LOW	lmdeploy/vl/model/preprocess_utils.py	189	def _expand_bundled_audio_items(item: dict, token_id: int) -> list[dict]:	CODE
LOW	lmdeploy/vl/model/llava.py	30	def _clip_vision_tower_load_model(self, **kwargs):	CODE
LOW	lmdeploy/vl/model/utils.py	11	def disable_transformers_logging():	CODE
LOW	lmdeploy/vl/model/deepseek.py	14	def check_deepseek_vl_install():	CODE
LOW	lmdeploy/vl/model/base.py	383	def to_pytorch_with_input_ids(self, messages):	CODE
LOW	lmdeploy/vl/media/video.py	46	def _get_video_loader_backend(self) -> VideoLoader:	CODE
LOW	lmdeploy/turbomind/tokenizer_info.py	119	def _is_sentencepiece_tokenizer(tokenizer: PreTrainedTokenizerBase) -> bool:	STRING
LOW	lmdeploy/turbomind/converter.py	95	def _validate_quant_group_size(model_format: str \| None, group_size: int \| None) -> int \| None:	CODE
LOW	lmdeploy/turbomind/turbomind.py	42	def _construct_stop_or_bad_words(words: list[int] = None):	CODE
LOW	lmdeploy/turbomind/turbomind.py	621	def _get_extra_output_processors(self, outputs: dict[str, torch.Tensor], gen_config: GenerationConfig,	CODE
LOW	lmdeploy/turbomind/models/internvl.py	122	def _validate_legacy_internvl_chat(cfg):	CODE
LOW	lmdeploy/turbomind/models/qwen2_vl.py	112	def _vision_intermediate_size(self, cfg):	CODE
2119 more matches not shown…

Decorative Section Separators302 hits · 1017 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	lmdeploy/deepseek_v4_encoding.py	97	# ============================================================	COMMENT
MEDIUM⚡	lmdeploy/deepseek_v4_encoding.py	99	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	14	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	16	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	39	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	41	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	223	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	225	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	405	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	407	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	510	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	512	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	609	# ============================================================	COMMENT
MEDIUM	lmdeploy/deepseek_v4_encoding.py	611	# ============================================================	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	29	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	31	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	67	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	69	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	104	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	106	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	221	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/linear.py	223	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	42	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	44	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	83	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	85	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	172	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	174	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	426	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/weight_format.py	428	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	10	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	12	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	26	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	28	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	45	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	47	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	116	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	118	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	216	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	218	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	234	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	236	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	250	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	252	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	350	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	352	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	365	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	367	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	422	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/_base.py	424	# ------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/attention.py	17	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/attention.py	19	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/attention.py	79	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/attention.py	81	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lmdeploy/turbomind/builders/ffn.py	26	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	lmdeploy/turbomind/builders/ffn.py	28	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/ffn.py	68	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/ffn.py	70	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/ffn.py	173	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	lmdeploy/turbomind/builders/ffn.py	175	# ---------------------------------------------------------------------------	COMMENT
242 more matches not shown…

Unused Imports582 hits · 573 pts

Severity	File	Line	Context
LOW	benchmark/benchmark_parser.py	9	CODE
LOW	benchmark/benchmark_parser.py	19	CODE
LOW	benchmark/benchmark_parser.py	22	CODE
LOW	benchmark/benchmark_chat_completion.py	14	CODE
LOW	benchmark/benchmark_generate.py	21	CODE
LOW	lmdeploy/__init__.py	3	CODE
LOW	lmdeploy/__init__.py	3	CODE
LOW	lmdeploy/__init__.py	3	CODE
LOW	lmdeploy/__init__.py	4	CODE
LOW	lmdeploy/__init__.py	4	CODE
LOW	lmdeploy/__init__.py	4	CODE
LOW	lmdeploy/__init__.py	4	CODE
LOW	lmdeploy/__init__.py	4	CODE
LOW	lmdeploy/__init__.py	5	CODE
LOW	lmdeploy/__init__.py	6	CODE
LOW	lmdeploy/__init__.py	7	CODE
LOW	lmdeploy/__init__.py	8	CODE
LOW	lmdeploy/__init__.py	8	CODE
LOW	lmdeploy/api.py	2	CODE
LOW	lmdeploy/utils.py	511	CODE
LOW	lmdeploy/utils.py	519	CODE
LOW	lmdeploy/utils.py	470	CODE
LOW	lmdeploy/pipeline.py	2	CODE
LOW	lmdeploy/lite/__init__.py	2	CODE
LOW	lmdeploy/lite/__init__.py	3	CODE
LOW	lmdeploy/lite/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	3	CODE
LOW	lmdeploy/lite/utils/__init__.py	3	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	4	CODE
LOW	lmdeploy/lite/utils/__init__.py	14	CODE
LOW	lmdeploy/lite/utils/__init__.py	15	CODE
LOW	lmdeploy/lite/utils/__init__.py	15	CODE
LOW	lmdeploy/lite/utils/__init__.py	15	CODE
LOW	lmdeploy/lite/utils/__init__.py	16	CODE
LOW	lmdeploy/lite/utils/__init__.py	17	CODE
LOW	lmdeploy/lite/quantization/__init__.py	2	CODE
LOW	lmdeploy/lite/quantization/__init__.py	2	CODE
LOW	lmdeploy/lite/quantization/__init__.py	3	CODE
LOW	lmdeploy/lite/quantization/__init__.py	3	CODE
LOW	lmdeploy/lite/quantization/__init__.py	4	CODE
LOW	lmdeploy/lite/quantization/weight/__init__.py	2	CODE
LOW	lmdeploy/lite/quantization/activation/__init__.py	2	CODE
LOW	lmdeploy/lite/quantization/activation/__init__.py	2	CODE
LOW	lmdeploy/lite/quantization/modules/__init__.py	2	CODE
LOW	lmdeploy/lite/model/__init__.py	3	CODE
LOW	lmdeploy/lite/model/__init__.py	3	CODE
LOW	lmdeploy/lite/model/__init__.py	7	CODE
LOW	lmdeploy/cli/__init__.py	2	CODE
LOW	lmdeploy/vl/__init__.py	2	CODE
LOW	lmdeploy/vl/__init__.py	2	CODE
LOW	lmdeploy/vl/__init__.py	2	CODE
LOW	lmdeploy/vl/__init__.py	2	CODE
LOW	lmdeploy/vl/__init__.py	2	CODE
522 more matches not shown…

Excessive Try-Catch Wrapping276 hits · 317 pts

Severity	File	Line	Snippet	Context
LOW	benchmark/benchmark_chat_completion.py	35	except Exception: # noqa: BLE001 - tqdm is optional for CLI progress display.	CODE
LOW	benchmark/benchmark_chat_completion.py	531	except Exception as e: # noqa: BLE001 - record and keep consuming SSE.	CODE
LOW	benchmark/benchmark_chat_completion.py	537	except Exception as e: # noqa: BLE001 - benchmark should record failures and continue.	CODE
LOW	benchmark/benchmark_chat_completion.py	772	except Exception as e:	CODE
LOW	benchmark/benchmark_pipeline.py	58	except Exception as e:	CODE
LOW	benchmark/benchmark_throughput.py	58	except Exception as e:	CODE
LOW	benchmark/benchmark_serving.py	87	except Exception as e:	CODE
LOW	benchmark/benchmark_serving.py	146	except Exception as e:	CODE
LOW	benchmark/benchmark_generate.py	56	except Exception: # noqa: BLE001	CODE
LOW	benchmark/benchmark_generate.py	274	except Exception as e: # noqa: BLE001	CODE
MEDIUM	benchmark/profile_restful_api.py	747	print(f'Error applying chat template: {e}, fallback to <image> tag')	CODE
MEDIUM	benchmark/profile_restful_api.py	1115	print(f'Error running benchmark for request rate: {request_rate}')	CODE
LOW	benchmark/profile_restful_api.py	148	except Exception:	CODE
LOW	benchmark/profile_restful_api.py	227	except Exception:	CODE
LOW	benchmark/profile_restful_api.py	335	except Exception:	CODE
LOW	benchmark/profile_restful_api.py	412	except Exception:	CODE
LOW	benchmark/profile_restful_api.py	744	except Exception as e:	CODE
LOW	benchmark/profile_restful_api.py	775	except Exception:	CODE
LOW	benchmark/profile_restful_api.py	1181	except Exception as e:	CODE
LOW	benchmark/profile_restful_api.py	1244	except Exception as e:	CODE
LOW⚡	lmdeploy/deepseek_v4_encoding.py	105	except Exception:	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	158	except Exception:	CODE
LOW	lmdeploy/model.py	653	except Exception:	CODE
LOW	lmdeploy/model.py	660	except Exception:	CODE
LOW	lmdeploy/model.py	707	except Exception:	CODE
LOW	lmdeploy/model.py	714	except Exception:	CODE
LOW	lmdeploy/model.py	777	except Exception as e:	CODE
LOW	lmdeploy/model.py	793	except Exception as e:	CODE
LOW	lmdeploy/model.py	864	except Exception:	CODE
LOW	lmdeploy/model.py	872	except Exception:	CODE
LOW	lmdeploy/utils.py	471	except Exception as e:	CODE
LOW	lmdeploy/utils.py	512	except Exception:	CODE
LOW	lmdeploy/utils.py	520	except Exception:	CODE
MEDIUM	lmdeploy/utils.py	509	def is_deep_ep_installed():	CODE
MEDIUM	lmdeploy/utils.py	517	def is_deep_gemm_installed():	CODE
LOW	lmdeploy/pipeline.py	408	except Exception:	CODE
LOW⚡	lmdeploy/deepseek_v32_encoding.py	67	except Exception:	CODE
LOW	lmdeploy/metrics/metrics_processor.py	81	except Exception as e:	CODE
LOW	lmdeploy/hf_configs/__init__.py	25	except Exception as error:	CODE
LOW	lmdeploy/hf_configs/__init__.py	41	except Exception:	CODE
LOW	lmdeploy/lite/apis/gptq.py	43	except Exception:	CODE
MEDIUM	lmdeploy/cli/chat.py	173	print(f'Error: {exc}')	CODE
LOW	lmdeploy/cli/cli.py	130	except Exception:	CODE
LOW	lmdeploy/vl/engine.py	69	except Exception as e:	CODE
LOW	lmdeploy/vl/model/builder.py	81	except Exception as e:	CODE
LOW	lmdeploy/vl/model/utils.py	44	except Exception:	CODE
LOW	lmdeploy/vl/model/base.py	101	except Exception as e:	CODE
LOW	lmdeploy/vl/media/time_series.py	42	except Exception as e:	CODE
LOW	lmdeploy/vl/media/connection.py	53	except Exception as e:	CODE
LOW	lmdeploy/turbomind/turbomind.py	818	except Exception as e:	CODE
LOW	lmdeploy/serve/parsers/_openai_harmony.py	128	except Exception:	CODE
LOW	…/serve/parsers/tool_parser/deepseek_v32_tool_parser.py	78	except Exception:	CODE
LOW	lmdeploy/serve/core/health.py	114	except Exception as e:	CODE
LOW	lmdeploy/serve/core/async_engine.py	316	except Exception:	CODE
LOW	lmdeploy/serve/core/async_engine.py	328	except Exception as e:	CODE
LOW	lmdeploy/serve/core/async_engine.py	453	except Exception:	CODE
LOW	lmdeploy/serve/core/async_engine.py	469	except Exception as e:	CODE
LOW	lmdeploy/serve/core/async_engine.py	606	except Exception as e:	CODE
LOW	lmdeploy/serve/proxy/streaming_response.py	54	except Exception:	CODE
LOW	lmdeploy/serve/proxy/proxy.py	200	except Exception as e: # noqa	CODE
216 more matches not shown…

Cross-File Repetition63 hits · 315 pts

Severity	File	Line	Snippet	Context
HIGH	lmdeploy/vl/model/molmo.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/llama4.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/llava.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/deepseek.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/llava_hf.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/llava_next.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/gemma3_vl.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/minicpmv.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/deepseek_vl2.py	0	build the vision part of a vlm model when backend is turbomind, or load the whole vlm model when `self.with_llm==true`	STRING
HIGH	lmdeploy/vl/model/molmo.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/llama4.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/llava.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/deepseek.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/llava_hf.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/llava_next.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/gemma3_vl.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/minicpmv.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/vl/model/deepseek_vl2.py	0	extract image feature. only implement it when the backend is turbomind engine. args: messages(list[dict]): the outputs o	STRING
HIGH	lmdeploy/serve/openai/protocol.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/engine.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base_worker.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/base_worker.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/ray_executor.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/base.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/serve/openai/endpoints/management.py	0	tear down a previously initialized weights-update group.	STRING
HIGH	lmdeploy/pytorch/messages.py	0	update token ids, old token ids will be added to history.	STRING
HIGH	lmdeploy/pytorch/strategies/dllm/sequence.py	0	update token ids, old token ids will be added to history.	STRING
HIGH	lmdeploy/pytorch/strategies/ar_spec/sequence.py	0	update token ids, old token ids will be added to history.	STRING
HIGH	lmdeploy/pytorch/strategies/ar/sequence.py	0	update token ids, old token ids will be added to history.	STRING
HIGH	lmdeploy/pytorch/strategies/dllm/step_inputs.py	0	shrink batch — keep only sequences at delta.indices.	STRING
HIGH	lmdeploy/pytorch/strategies/ar_spec/step_inputs.py	0	shrink batch — keep only sequences at delta.indices.	STRING
HIGH	lmdeploy/pytorch/strategies/ar/step_inputs.py	0	shrink batch — keep only sequences at delta.indices.	STRING
HIGH	lmdeploy/pytorch/strategies/base/step_inputs.py	0	shrink batch — keep only sequences at delta.indices.	STRING
HIGH	lmdeploy/pytorch/backends/cuda/attention/fa3.py	0	describe metadata required by this selected implementation.	STRING
HIGH	lmdeploy/pytorch/backends/cuda/attention/default.py	0	describe metadata required by this selected implementation.	STRING
HIGH	lmdeploy/pytorch/backends/cuda/attention/mla.py	0	describe metadata required by this selected implementation.	STRING
HIGH	lmdeploy/pytorch/models/deepseek_v4.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/kimi_k2_language.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/qwen3_5.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/qwen3_5_moe.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/interns2_mobius.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/meta_moe.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/qwen3_next.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	lmdeploy/pytorch/models/deepseek_v2.py	0	we might change the number of layers so we can debug the model with less gpus.	STRING
HIGH	…/pytorch/paging/block_manager/default_block_manager.py	0	return if physical block can be allocated for given message.	STRING
HIGH	…y/pytorch/paging/block_manager/window_block_manager.py	0	return if physical block can be allocated for given message.	STRING
HIGH	…loy/pytorch/paging/block_manager/base_block_manager.py	0	return if physical block can be allocated for given message.	STRING
HIGH	…/pytorch/paging/block_manager/default_block_manager.py	0	allocate physical blocks for given message according to logical blocks.	STRING
HIGH	…y/pytorch/paging/block_manager/window_block_manager.py	0	allocate physical blocks for given message according to logical blocks.	STRING
HIGH	…loy/pytorch/paging/block_manager/base_block_manager.py	0	allocate physical blocks for given message according to logical blocks.	STRING
HIGH	lmdeploy/pytorch/engine/engine.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base_worker.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/base_worker.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/ray_executor.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/executor/base.py	0	receive weights through the disaggregated process group.	STRING
HIGH	lmdeploy/pytorch/engine/base.py	0	drop connection. 1. drop engine connection (zmq connection) 2. todo(jimyma) drop rdma connection.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base_worker.py	0	drop connection. 1. drop engine connection (zmq connection) 2. todo(jimyma) drop rdma connection.	STRING
HIGH	lmdeploy/pytorch/engine/mp_engine/base.py	0	drop connection. 1. drop engine connection (zmq connection) 2. todo(jimyma) drop rdma connection.	STRING
3 more matches not shown…

Magic Placeholder Names58 hits · 288 pts

Severity	File	Line	Snippet	Context
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	643	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	659	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	680	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	699	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	723	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	745	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	772	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	794	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	822	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	837	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	853	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	872	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	898	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	914	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	931	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	948	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	979	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	1000	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	1031	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	…/interface/restful/test_restful_chat_completions_v1.py	1064	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{BASE_URL}/v1')	CODE
HIGH	autotest/utils/anthropic_messages.py	380	api_key=os.getenv('ANTHROPIC_API_KEY', 'YOUR_API_KEY'),	CODE
HIGH	autotest/utils/tool_reasoning_definitions.py	206	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{url}/v1')	CODE
HIGH	autotest/utils/tool_reasoning_definitions.py	949	'Authorization': 'Bearer YOUR_API_KEY',	CODE
HIGH	autotest/utils/tool_reasoning_definitions.py	1106	'Authorization': 'Bearer YOUR_API_KEY',	CODE
HIGH	autotest/utils/run_restful_chat.py	181	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{url}/v1')	CODE
HIGH	autotest/utils/run_restful_chat.py	510	client = OpenAI(api_key='YOUR_API_KEY', base_url=http_url + '/v1')	CODE
HIGH	autotest/utils/run_restful_chat.py	851	client = OpenAI(api_key='YOUR_API_KEY', base_url=http_url + '/v1')	CODE
HIGH	autotest/utils/run_restful_chat.py	1142	client = OpenAI(api_key='YOUR_API_KEY', base_url=http_url + '/v1')	CODE
HIGH	docs/zh_cn/llm/api_server.md	90	api_key='YOUR_API_KEY',	CODE
HIGH	docs/zh_cn/llm/api_server.md	113	client = AsyncOpenAI(api_key='YOUR_API_KEY',	CODE
HIGH	docs/zh_cn/llm/codellama.md	149	api_key='YOUR_API_KEY',	STRING
HIGH	docs/zh_cn/llm/api_server_tools.md	39	client = OpenAI(api_key='YOUR_API_KEY',base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/zh_cn/llm/api_server_tools.md	112	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/zh_cn/llm/api_server_tools.md	186	client = OpenAI(api_key='YOUR_API_KEY',	CODE
HIGH	docs/zh_cn/llm/api_server_tools.md	357	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/zh_cn/llm/api_server_lora.md	77	api_key='YOUR_API_KEY',	CODE
HIGH	docs/zh_cn/advance/structed_output.md	99	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/zh_cn/advance/spec_decoding.md	186	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:24545/v1')	CODE
HIGH	docs/zh_cn/get_started/get_started.md	131	api_key='YOUR_API_KEY',	CODE
HIGH	docs/zh_cn/get_started/get_started.md	165	client = OpenAI(api_key='YOUR_API_KEY', # A dummy api_key is required	CODE
HIGH	docs/zh_cn/multi_modal/api_server_vl.md	60	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/zh_cn/multi_modal/phi3.md	55	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/llm/api_server.md	71	api_key='YOUR_API_KEY',	CODE
HIGH	docs/en/llm/api_server.md	94	client = AsyncOpenAI(api_key='YOUR_API_KEY',	CODE
HIGH	docs/en/llm/codellama.md	149	api_key='YOUR_API_KEY',	STRING
HIGH	docs/en/llm/api_server_tools.md	39	client = OpenAI(api_key='YOUR_API_KEY',base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/llm/api_server_tools.md	112	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/llm/api_server_tools.md	186	client = OpenAI(api_key='YOUR_API_KEY',	CODE
HIGH	docs/en/llm/api_server_tools.md	357	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/llm/api_server_lora.md	77	api_key='YOUR_API_KEY',	CODE
HIGH	docs/en/advance/structed_output.md	99	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/advance/spec_decoding.md	187	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:24545/v1')	CODE
HIGH	docs/en/get_started/get_started.md	134	api_key='YOUR_API_KEY',	CODE
HIGH	docs/en/get_started/get_started.md	169	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/multi_modal/api_server_vl.md	83	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	docs/en/multi_modal/phi3.md	56	client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')	CODE
HIGH	eval/config.py	57	key='YOUR_API_KEY',	CODE
HIGH⚡	eval/eval.py	121	client = OpenAI(api_key='YOUR_API_KEY', base_url=f'{server}/v1')	CODE

Deep Nesting263 hits · 248 pts

Severity	File	Line	Context
LOW	setup.py	63	CODE
LOW	setup.py	79	CODE
LOW	setup.py	120	CODE
LOW	benchmark/benchmark_parser.py	81	CODE
LOW	benchmark/benchmark_chat_completion.py	266	CODE
LOW	benchmark/benchmark_chat_completion.py	448	CODE
LOW	benchmark/benchmark_chat_completion.py	1005	CODE
LOW	benchmark/benchmark_guided.py	527	CODE
LOW	benchmark/benchmark_generate.py	203	CODE
LOW	benchmark/benchmark_generate.py	280	CODE
LOW	benchmark/profile_restful_api.py	159	CODE
LOW	benchmark/profile_restful_api.py	237	CODE
LOW	benchmark/profile_restful_api.py	345	CODE
LOW	benchmark/profile_restful_api.py	1186	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	227	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	409	CODE
LOW	lmdeploy/deepseek_v4_encoding.py	469	CODE
LOW	lmdeploy/model.py	649	CODE
LOW	lmdeploy/model.py	703	CODE
LOW	lmdeploy/tokenizer.py	152	CODE
LOW	lmdeploy/utils.py	394	CODE
LOW	lmdeploy/utils.py	420	CODE
LOW	lmdeploy/utils.py	542	CODE
LOW	lmdeploy/deepseek_v32_encoding.py	143	CODE
LOW	lmdeploy/archs.py	54	CODE
LOW	lmdeploy/archs.py	93	CODE
LOW	lmdeploy/archs.py	166	CODE
LOW	lmdeploy/metrics/metrics_processor.py	45	CODE
LOW	lmdeploy/metrics/stats.py	146	CODE
LOW	lmdeploy/lite/apis/auto_awq.py	18	CODE
LOW	lmdeploy/lite/apis/smooth_quant.py	18	CODE
LOW	lmdeploy/lite/apis/calibrate.py	201	CODE
LOW	lmdeploy/lite/apis/calibrate.py	233	CODE
LOW	lmdeploy/lite/utils/calib_dataloader.py	14	CODE
LOW	lmdeploy/lite/utils/memory_efficient.py	197	CODE
LOW	lmdeploy/lite/utils/batch_split.py	7	CODE
LOW	lmdeploy/cli/entrypoint.py	10	CODE
LOW	lmdeploy/cli/chat.py	146	CODE
LOW	lmdeploy/cli/utils.py	14	CODE
LOW	lmdeploy/cli/utils.py	856	CODE
LOW	lmdeploy/vl/model/internvl.py	21	CODE
LOW	lmdeploy/vl/model/internvl.py	173	CODE
LOW	lmdeploy/vl/model/molmo.py	129	CODE
LOW	lmdeploy/vl/model/molmo.py	151	CODE
LOW	lmdeploy/vl/model/preprocess_utils.py	207	CODE
LOW	lmdeploy/vl/model/llava.py	305	CODE
LOW	lmdeploy/vl/model/builder.py	36	CODE
LOW	lmdeploy/vl/model/utils.py	29	CODE
LOW	lmdeploy/vl/model/qwen2.py	47	CODE
LOW	lmdeploy/vl/model/base.py	123	CODE
LOW	lmdeploy/vl/model/base.py	363	CODE
LOW	lmdeploy/vl/model/minicpmv.py	200	CODE
LOW	lmdeploy/vl/media/time_series.py	31	CODE
LOW	lmdeploy/turbomind/checkpoint.py	188	CODE
LOW	lmdeploy/turbomind/tokenizer_info.py	133	CODE
LOW	lmdeploy/turbomind/linear.py	109	CODE
LOW	lmdeploy/turbomind/linear.py	160	CODE
LOW	lmdeploy/turbomind/linear.py	121	CODE
LOW	lmdeploy/turbomind/linear.py	171	CODE
LOW	lmdeploy/turbomind/converter.py	29	CODE
203 more matches not shown…

AI Structural Patterns153 hits · 150 pts

Severity	File	Line	Context
LOW	benchmark/profile_restful_api.py	1166	CODE
LOW	lmdeploy/model.py	117	CODE
LOW	lmdeploy/model.py	204	CODE
LOW	lmdeploy/model.py	726	CODE
LOW	lmdeploy/api.py	86	CODE
LOW	lmdeploy/utils.py	638	CODE
LOW	lmdeploy/lite/apis/auto_awq.py	41	CODE
LOW	lmdeploy/lite/apis/smooth_quant.py	18	CODE
LOW	lmdeploy/lite/apis/gptq.py	11	CODE
LOW	lmdeploy/lite/apis/calibrate.py	283	CODE
LOW	lmdeploy/lite/utils/memory_efficient.py	124	CODE
LOW	lmdeploy/vl/model/base.py	238	CODE
LOW	lmdeploy/turbomind/models/utils.py	225	CODE
LOW	lmdeploy/serve/core/async_engine.py	119	CODE
LOW	lmdeploy/serve/core/async_engine.py	478	CODE
LOW	lmdeploy/serve/proxy/proxy.py	889	CODE
LOW	lmdeploy/serve/proxy/proxy.py	325	CODE
LOW	lmdeploy/serve/openai/api_server.py	257	CODE
LOW	lmdeploy/serve/openai/api_client.py	90	CODE
LOW	lmdeploy/serve/openai/api_client.py	175	CODE
LOW	lmdeploy/pytorch/config.py	519	CODE
LOW	lmdeploy/pytorch/model_inputs.py	206	CODE
LOW	lmdeploy/pytorch/model_inputs.py	300	CODE
LOW	lmdeploy/pytorch/model_inputs.py	438	CODE
LOW	lmdeploy/pytorch/messages.py	368	CODE
LOW	lmdeploy/pytorch/messages.py	485	CODE
LOW	lmdeploy/pytorch/kernels/cuda/pagedattention.py	772	CODE
LOW	lmdeploy/pytorch/kernels/cuda/v4_flatten_kv.py	201	CODE
LOW	lmdeploy/pytorch/kernels/cuda/fused_lora.py	159	CODE
LOW	lmdeploy/pytorch/kernels/cuda/flashattention.py	463	CODE
LOW	lmdeploy/pytorch/kernels/cuda/utils.py	46	CODE
LOW	…eploy/pytorch/kernels/cuda/compressed_tensors_w4a16.py	479	CODE
LOW	lmdeploy/pytorch/kernels/cuda/gated_delta_rule.py	256	CODE
LOW	lmdeploy/pytorch/kernels/cuda/gated_delta_rule.py	575	CODE
LOW	lmdeploy/pytorch/kernels/cuda/w8a8_triton_kernels.py	379	CODE
LOW	lmdeploy/pytorch/kernels/cuda/causal_conv1d.py	212	CODE
LOW	lmdeploy/pytorch/kernels/cuda/rms_norm.py	18	CODE
LOW	lmdeploy/pytorch/kernels/cuda/flatten_kv_cache.py	397	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/blocked_fp8.py	160	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/blocked_fp8.py	396	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/blocked_fp8.py	820	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/fused_moe.py	545	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/fused_moe.py	1137	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/w8a8.py	341	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/v4_fp4.py	228	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/v4_fp4.py	425	CODE
LOW	lmdeploy/pytorch/kernels/cuda/moe/v4_fp4.py	341	CODE
LOW	lmdeploy/pytorch/nn/attention.py	24	CODE
LOW	lmdeploy/pytorch/nn/norm.py	31	CODE
LOW	lmdeploy/pytorch/nn/rotary_embedding.py	177	CODE
LOW	lmdeploy/pytorch/nn/nsa.py	47	CODE
LOW	lmdeploy/pytorch/nn/linear/blocked_fp8.py	19	CODE
LOW	lmdeploy/pytorch/nn/linear/blocked_fp8.py	169	CODE
LOW	lmdeploy/pytorch/nn/linear/blocked_fp8.py	277	CODE
LOW	lmdeploy/pytorch/nn/linear/__init__.py	29	CODE
LOW	lmdeploy/pytorch/nn/linear/__init__.py	128	CODE
LOW	lmdeploy/pytorch/nn/linear/__init__.py	172	CODE
LOW	lmdeploy/pytorch/nn/linear/__init__.py	209	CODE
LOW	lmdeploy/pytorch/nn/linear/__init__.py	296	CODE
LOW	lmdeploy/pytorch/nn/linear/default.py	18	CODE
93 more matches not shown…

Over-Commented Block120 hits · 120 pts

Severity	File	Line	Snippet	Context
LOW	CMakeLists.txt	1	# Copyright (c) 2019-2023, NVIDIA CORPORATION. All rights reserved.	COMMENT
LOW	benchmark/benchmark_decode.py	61	if __name__ == '__main__':	COMMENT
LOW	benchmark/benchmark_decode.py	81	# Decoded 15991314 tokens in 242.7 seconds, 65893.38488718234 tokens/s.	COMMENT
LOW	lmdeploy/pytorch/utils.py	161	# When using `remote_code` in HF components like tokenizer or config	COMMENT
LOW	lmdeploy/pytorch/kernels/cuda/v4_compressor.py	1121		COMMENT
LOW	lmdeploy/pytorch/kernels/cuda/flashattention.py	141	k1_ptrs = tl.advance(k1_ptrs, (0, BLOCK_N))	COMMENT
LOW	lmdeploy/pytorch/strategies/dllm/step_inputs.py	1	# Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	lmdeploy/pytorch/strategies/ar_spec/step_inputs.py	1	# Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	lmdeploy/pytorch/strategies/ar/step_inputs.py	1	# Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	lmdeploy/pytorch/models/kimi_k25_vision.py	1	# Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	lmdeploy/pytorch/models/kimi_k25_vision.py	21	# See the License for the specific language governing permissions and	COMMENT
LOW	tests/test_lmdeploy/test_model.py	281	# @pytest.mark.parametrize('has_user_sys', [True, False])	COMMENT
LOW	tests/test_lmdeploy/test_model.py	301	# 'role': 'user',	COMMENT
LOW	tests/pytorch/test_internvl_lora.py	21		COMMENT
LOW	docs/zh_cn/conf.py	1	#	COMMENT
LOW	docs/zh_cn/conf.py	121	master_doc = 'index'	COMMENT
LOW	docs/zh_cn/conf.py	161	'use_edit_page_button': True,	COMMENT
LOW	docs/zh_cn/conf.py	201	#	COMMENT
LOW	docs/zh_cn/conf.py	221	# The font size ('10pt', '11pt' or '12pt').	COMMENT
LOW	docs/en/conf.py	1	#	COMMENT
LOW	docs/en/conf.py	121	master_doc = 'index'	COMMENT
LOW	docs/en/conf.py	161	'use_source_button': True,	COMMENT
LOW	docs/en/conf.py	201	# The default sidebars (for documents that don't match any pattern) are	COMMENT
LOW	docs/en/conf.py	221	#	COMMENT
LOW	src/CMakeLists.txt	1	# Copyright (c) 2019-2023, NVIDIA CORPORATION. All rights reserved.	COMMENT
LOW	src/turbomind/turbomind.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/CMakeLists.txt	1	# Copyright (c) 2019-2023, NVIDIA CORPORATION. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/CMakeLists.txt	1	# Copyright (c) 2019-2023, NVIDIA CORPORATION. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/attention/impl_simt.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/attention/iterator_sm80.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/attention/impl_884.h	41	static constexpr int V_N = HeadDim / OP_N; // 8	COMMENT
LOW	src/turbomind/kernels/core/floating_point.h	141	// FloatingPoint<5, 2>::max_denormal;	COMMENT
LOW	src/turbomind/kernels/core/common.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/core/common.h	21	#endif	COMMENT
LOW	src/turbomind/kernels/core/common.h	41	#if defined(__CUDACC_RTC__) \|\| (defined(__clang__) && defined(__CUDA__))	COMMENT
LOW	src/turbomind/kernels/core/common.h	61	#define TM_HOST __forceinline__ __host__	COMMENT
LOW	src/turbomind/kernels/gemm/iterator_sm70.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/tiled_mma.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/kernel_impl.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/moe_gate_python_bind.cpp	1	// src/turbomind/kernels/gemm/moe_gate_python_bind.cpp	COMMENT
LOW	src/turbomind/kernels/gemm/gmma_fp8_sm90.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_v3.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_v3.h	21	#include "cutlass/pipeline/sm90_pipeline.hpp"	COMMENT
LOW	src/turbomind/kernels/gemm/mainloop_sm80_v2.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/iterator_sm80.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	…urbomind/kernels/gemm/prepare_moe_tma_descs_sm90_fp8.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/matrix_ptr.h	101		COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_bf16.h	61	#include "cute/arch/cluster_sm90.hpp"	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_bf16.h	81	#include "src/turbomind/kernels/gemm/cp_async.h"	COMMENT
LOW	src/turbomind/kernels/gemm/moe_utils_v2.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/smem_copy.h	181	}	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_fp8_wa.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_fp8_wa.h	21	#include "cutlass/pipeline/sm90_pipeline.hpp"	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_v5.h	1	#pragma once	COMMENT
LOW	src/turbomind/kernels/gemm/gemm_universal_sm90_v5.h	21		COMMENT
LOW	src/turbomind/kernels/gemm/mainloop_sm70.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/kernel_impl_sm90_bf16.h	1	// Copyright (c) OpenMMLab. All rights reserved.	COMMENT
LOW	src/turbomind/kernels/gemm/sm90_bf16_traits.h	1	#pragma once	COMMENT
60 more matches not shown…

Modern Structural Boilerplate58 hits · 58 pts

Severity	File	Line	Snippet	Context
LOW	lmdeploy/__init__.py	10	__all__ = [	CODE
LOW	lmdeploy/lite/utils/__init__.py	19	__all__ = [	CODE
LOW	lmdeploy/lite/quantization/__init__.py	6	__all__ = ['WeightQuantizer', 'ActivationObserver', 'KVCacheObserver', 'CalibrationContext', 'CalibrationContextV2']	CODE
LOW	lmdeploy/lite/quantization/weight/__init__.py	4	__all__ = ['WeightQuantizer']	CODE
LOW	lmdeploy/lite/quantization/activation/__init__.py	4	__all__ = ['ActivationObserver', 'KVCacheObserver']	CODE
LOW	lmdeploy/lite/quantization/modules/__init__.py	4	__all__ = ['WeightOnlyQLinear']	CODE
LOW	lmdeploy/lite/model/__init__.py	9	__all__ = ['MODELS']	CODE
LOW	lmdeploy/cli/__init__.py	4	__all__ = ['run']	CODE
LOW	lmdeploy/vl/__init__.py	13	__all__ = [	CODE
LOW	lmdeploy/turbomind/tokenizer_info.py	23	logger = logging.getLogger(__name__)	CODE
LOW	lmdeploy/turbomind/__init__.py	54	__all__ = ['TurboMind', 'update_parallel_config']	CODE
LOW⚡	lmdeploy/turbomind/builders/ffn.py	18	__all__ = [	CODE
LOW	lmdeploy/turbomind/builders/__init__.py	17	__all__ = [	CODE
LOW	lmdeploy/serve/__init__.py	6	__all__ = [	CODE
LOW	lmdeploy/serve/parsers/__init__.py	6	__all__ = ['ResponseParser', 'ResponseParserManager', 'GptOssResponseParser', 'validate_parser_names']	CODE
LOW	lmdeploy/serve/parsers/response_parser.py	654	def _update_reasoning_tokens(self, token_ids: list[int]) -> None:	CODE
LOW	lmdeploy/serve/parsers/tool_parser/__init__.py	15	__all__ = [	CODE
LOW	lmdeploy/serve/parsers/reasoning_parser/__init__.py	7	__all__ = [	CODE
LOW	lmdeploy/serve/core/__init__.py	7	__all__ = [	CODE
LOW	lmdeploy/serve/managers/__init__.py	4	__all__ = ['Session', 'SessionManager']	CODE
LOW	lmdeploy/serve/processors/__init__.py	4	__all__ = ['MultimodalProcessor']	CODE
LOW	lmdeploy/serve/anthropic/__init__.py	6	__all__ = ['create_anthropic_router']	CODE
LOW	lmdeploy/serve/openai/endpoints/__init__.py	7	__all__ = ['create_openai_router']	CODE
LOW	lmdeploy/serve/openai/responses/__init__.py	13	__all__ = [	CODE
LOW	lmdeploy/serve/openai/chat_completions/__init__.py	13	__all__ = ['register']	CODE
LOW	lmdeploy/pytorch/prefix_cache_state.py	134	def set_fresh_block_range(self, start_block_idx: int, end_block_idx: int) -> None:	CODE
LOW	lmdeploy/pytorch/configurations/__init__.py	7	__all__ = []	CODE
LOW	lmdeploy/pytorch/kernels/__init__.py	10	__all__ = [	CODE
LOW	lmdeploy/pytorch/kernels/cuda/__init__.py	14	__all__ = [	CODE
LOW	lmdeploy/pytorch/kernels/cuda/turbo_quant.py	14	logger = logging.getLogger(__name__)	CODE
LOW	lmdeploy/pytorch/kernels/default/__init__.py	5	__all__ = [	CODE
LOW	lmdeploy/pytorch/kernels/dlinfer/__init__.py	13	__all__ = [	CODE
LOW	lmdeploy/pytorch/kv_connector/__init__.py	15	__all__ = [	CODE
LOW	lmdeploy/pytorch/kv_connector/mooncake/store/worker.py	231	def _setup_store(self, store: Any, local_hostname: str) -> None:	CODE
LOW	…deploy/pytorch/kv_connector/mooncake/store/protocol.py	11	__all__ = ['LOOKUP_MSG', 'RESP_ERR']	CODE
LOW	lmdeploy/pytorch/kv_connector/mooncake/store/lookup.py	366	__all__ = ['LookupKeyClient', 'LookupKeyServer', 'prepare_lookup_rpc_path']	CODE
LOW	…deploy/pytorch/kv_connector/mooncake/store/__init__.py	6	__all__ = [	CODE
LOW	lmdeploy/pytorch/backends/cuda/attention/fa3.py	130	def update_fa3_metadata(attn_metadata, step_context, **kwargs) -> None:	CODE
LOW	lmdeploy/pytorch/memdecode/__init__.py	6	__all__ = [	CODE
LOW	lmdeploy/pytorch/models/__init__.py	4	__all__ = ['QLinear', 'QRMSNorm']	CODE
LOW	lmdeploy/pytorch/models/kimi_k25_vision.py	809	__all__ = [	CODE
LOW	lmdeploy/pytorch/disagg/backend/__init__.py	18	__all__ = ['DLSlimeBackend', 'MooncakeBackend']	CODE
LOW	lmdeploy/pytorch/spec_decode/__init__.py	66	__all__ = ['build_spec_agent']	CODE
LOW	lmdeploy/pytorch/multimodal/__init__.py	4	__all__ = ['MultiModalData']	CODE
LOW	lmdeploy/pytorch/paging/__init__.py	4	__all__ = ['Scheduler']	CODE
LOW	lmdeploy/pytorch/paging/scheduler.py	1446	def update_connector_output(self, connector_output) -> None:	CODE
LOW	lmdeploy/pytorch/paging/block_trie/__init__.py	7	__all__ = ['BlockTrie', 'Node', 'PrefixCacheStats']	CODE
LOW	lmdeploy/pytorch/transformers/__init__.py	4	__all__ = ['config_from_pretrained', 'register_config']	CODE
LOW	lmdeploy/pytorch/engine/__init__.py	5	__all__ = ['Engine', 'EngineInstance']	CODE
LOW	lmdeploy/pytorch/engine/executor/base.py	442	def update_configs(self) -> None:	CODE
LOW	lmdeploy/pytorch/devices/__init__.py	4	__all__ = ['DeviceContext', 'DefaultContext', 'get_device_manager']	CODE
LOW	tests/turbomind/linear/linear.py	258	def set_grouped(self, grouped: bool) -> None:	CODE
LOW	tests/turbomind/linear/linear.py	266	def set_epilogue(self, epilogue) -> None:	CODE
LOW	tests/turbomind/linear/linear.py	299	def set_param(self, name: str, tensor) -> None:	CODE
LOW	tests/turbomind/linear/linear.py	399	def set_measure(self, on: bool) -> None:	CODE
LOW	tests/turbomind/linear_attn/cases.py	179	def set_logical(self, layer: int, state: torch.Tensor) -> None:	CODE
LOW	autotest/utils/config_utils.py	1488	def set_device_env_variable(worker_id: str \| None, parallel_config: dict[str, int] \| None = None) -> None:	CODE
LOW	scripts/test_turbomind_model.py	183	def _set_hf_cache(path: str) -> None:	CODE

Redundant / Tautological Comments14 hits · 23 pts

Severity	File	Line	Snippet	Context
LOW	benchmark/profile_restful_api.py	307	# Check if this chunk contains content	COMMENT
LOW	benchmark/profile_restful_api.py	519	# Check if the cache file already exists	COMMENT
LOW	lmdeploy/deepseek_v4_encoding.py	349	# Check if previous message has a task - if so, this is a task output (no thinking)	COMMENT
LOW	lmdeploy/model.py	781	# Verify if the model can perform apply_chat_template with different roles.	COMMENT
LOW⚡	lmdeploy/lite/apis/calibrate.py	153	# Check if the child is an instance of the given layer type	COMMENT
LOW⚡	lmdeploy/lite/apis/calibrate.py	161	# Check if the child contains the target module type	COMMENT
LOW⚡	lmdeploy/lite/apis/calibrate.py	164	# Check if the child matches the head name	COMMENT
LOW	lmdeploy/lite/utils/batch_split.py	102	# Check if the current element is a past key-value pair.	COMMENT
LOW	lmdeploy/serve/processors/multimodal.py	253	# Check if multimodal input exists	COMMENT
LOW	…loy/serve/openai/chat_completions/logits_processors.py	30	# Check if token_id is within the vocab size	COMMENT
LOW	lmdeploy/pytorch/engine/executor/ray_executor.py	646	# Check if all elements in ips are present in worker_ips and vice versa (ignoring order)	COMMENT
LOW	autotest/utils/run_restful_chat.py	114	# Check if process is still running	COMMENT
LOW	autotest/utils/run_restful_chat.py	1290	# Check if process is still running	COMMENT
LOW	autotest/utils/ray_distributed_utils.py	58	# Check if port is open	COMMENT

Self-Referential Comments11 hits · 19 pts

Severity	File	Line	Snippet	Context
MEDIUM	lmdeploy/api.py	97	"""This function is deprecated and no longer available.	STRING
MEDIUM	lmdeploy/api.py	110	"""This function is deprecated and no longer available.	STRING
MEDIUM	lmdeploy/lite/quantization/modules/linear.py	16	"""This class implements weight only quantization linear.	STRING
MEDIUM	lmdeploy/turbomind/tokenizer_info.py	3	"""This module provides the tokenizer info class to handle the tokenizer	STRING
MEDIUM	lmdeploy/pytorch/ray.py	127	# Create a new placement group	COMMENT
MEDIUM	lmdeploy/pytorch/tools/utils.py	165	# Create a clean table-like format	COMMENT
MEDIUM	lmdeploy/pytorch/models/llama4.py	751	"""This function is used to fetch the first embedding layer to activate	STRING
MEDIUM	lmdeploy/pytorch/models/kimi_k25_vision.py	3	# This file contains code adapted from moonshotai/Kimi-K2.6	COMMENT
MEDIUM	lmdeploy/pytorch/models/gemma3_vl.py	240	# Create a global causal mask.	STRING
MEDIUM	lmdeploy/pytorch/models/gemma3_vl.py	261	# Create a local causal mask with sliding window (1024).	STRING
MEDIUM	…_lite/test_quantization/test_utils/test_cal_qparams.py	18	# Create a dummy tensor	COMMENT

Structural Annotation Overuse8 hits · 13 pts

Severity	File	Line	Snippet	Context
LOW	lmdeploy/deepseek_v32_encoding.py	355	# NOTE: This function parses only correctly formatted strings and will not attempt to correct	COMMENT
LOW⚡	…rve/openai/chat_completions/test_delta_tool_call_id.py	18	# Step 1: feed partial JSON with name	COMMENT
LOW⚡	…rve/openai/chat_completions/test_delta_tool_call_id.py	27	# Step 2: feed final chunk with arguments	COMMENT
LOW⚡	…ts/pytorch/spec_decode/test_guided_spec_integration.py	481	# Step 2: original should be at correct state	COMMENT
LOW	docs/en/advance/update_weights.md	5	## Step 1: Launch server	COMMENT
LOW	docs/en/advance/update_weights.md	13	## Step 2: Offloads weights & kv cache	COMMENT
LOW	docs/en/advance/update_weights.md	38	## Step 3: Update weights	COMMENT
LOW	docs/en/advance/update_weights.md	71	## Step 4: Wakeup server	COMMENT

Docstring Block Structure2 hits · 10 pts

Severity	File	Line	Snippet	Context
HIGH	lmdeploy/api.py	24	Create a pipeline for inference. Args: model_path: the path of a model. It could be one of the following op	STRING
HIGH	lmdeploy/lite/utils/memory_efficient.py	46	Finds modules in model that return given value. Args: model (nn.Module): Model to inspect value (st	STRING

Dead Code4 hits · 8 pts

Severity	File	Line	Context
MEDIUM	…oy/serve/openai/chat_completions/test_n_completions.py	331	CODE
MEDIUM	…oy/serve/openai/chat_completions/test_n_completions.py	418	CODE
MEDIUM	…oy/serve/openai/chat_completions/test_n_completions.py	475	CODE
MEDIUM	tests/pytorch/engine/test_ray_mp_engine.py	166	CODE

Verbosity Indicators4 hits · 6 pts

Severity	File	Line	Snippet	Context
LOW	lmdeploy/serve/core/async_engine.py	809	The purpose of this function is to allow users to choose whether to use the synchronous interface or the	STRING
LOW⚡	…rve/openai/chat_completions/test_delta_tool_call_id.py	18	# Step 1: feed partial JSON with name	COMMENT
LOW⚡	…rve/openai/chat_completions/test_delta_tool_call_id.py	27	# Step 2: feed final chunk with arguments	COMMENT
LOW⚡	…ts/pytorch/spec_decode/test_guided_spec_integration.py	481	# Step 2: original should be at correct state	COMMENT

AI Slop Vocabulary3 hits · 6 pts

Severity	File	Line	Snippet	Context
MEDIUM	lmdeploy/pytorch/kernels/cuda/moe/blocked_fp8.py	645	# tokens. A compact BM64 schedule avoids that cliff and remains robust	COMMENT
LOW	lmdeploy/pytorch/paging/scheduler.py	1250	# No need to evict, just return True.	COMMENT
LOW	lmdeploy/pytorch/paging/scheduler.py	1253	# Enough free blocks, just return True.	COMMENT

Modern AI Meta-Vocabulary2 hits · 6 pts

Severity	File	Line	Snippet	Context
MEDIUM	autotest/tools/pipeline/mllm_case.py	374	# In-context few-shot learning	COMMENT
MEDIUM	docs/en/multi_modal/vl_pipeline.md	66	### Set context window size	COMMENT

Cross-Language Confusion2 hits · 5 pts

Severity	File	Line	Snippet	Context
HIGH	lmdeploy/turbomind/parallel_config.py	8	attn_tp == tp \|\| mlp_tp == tp	STRING
HIGH	lmdeploy/turbomind/parallel_config.py	9	tp % attn_tp == 0 && tp % mlp_tp == 0	STRING

Example Usage Blocks1 hit · 2 pts

Severity	File	Line	Snippet	Context
LOW	tests/pytorch/engine/test_zmq_rpc.py	76	# Example usage	COMMENT

Fake / Example Data1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	lmdeploy/model.py	844	messages = [{'role': 'user', 'content': 'placeholder'}, {'role': 'assistant', 'content': 'sentinel'}]	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings