mudler/LocalAI

5.0

Adjusted Score

5.0

Raw Score

100%

Time Factor

2026-08-28

Last Push

48.7K

Stars

Language

641.8K

Lines of Code

3.3K

Files

2.2K

Pattern Hits

2026-08-29

Scan Date

0.04

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 3HIGH 129MEDIUM 230LOW 1878

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 2240 distinct pattern matches across 24 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.

Over-Commented Block1066 hits · 1050 pts

Severity	File	Line	Snippet	Context
LOW	docker-compose.yaml	21	# Avoid probing remote gallery GGUF metadata during container startup.	COMMENT
LOW	docker-compose.yaml	41	# Here we can specify a list of models to run (see quickstart https://localai.io/basics/getting_started/#running-mod	COMMENT
LOW	docker-compose.yaml	61	# For legacy NVIDIA driver (for older NVIDIA Container Toolkit):	COMMENT
LOW	docker-compose.distributed.yaml	1	# Docker Compose for LocalAI Distributed Mode	COMMENT
LOW	docker-compose.distributed.yaml	61	# models directory at the SAME path (see "Shared Volume Mode" below).	COMMENT
LOW	docker-compose.distributed.yaml	141	# init: true	COMMENT
LOW	docker-compose.distributed.yaml	161	# backend-llama-cpp:	COMMENT
LOW	docker-compose.distributed.yaml	181	# will dynamically install the required backend via NATS when	COMMENT
LOW	cmd/launcher/internal/release_manager.go	41	type ReleaseManager struct {	COMMENT
LOW	.docker/llama-cpp-compile.sh	21	BUILD_TARGET=$(/LocalAI/.docker/llama-cpp-build-target.sh "${TARGETARCH}" "${BUILD_TYPE:-}")	COMMENT
LOW	.docker/llama-cpp-build-target.sh	1	#!/usr/bin/env bash	COMMENT
LOW	.docker/install-base-deps.sh	1	#!/usr/bin/env bash	COMMENT
LOW	.docker/install-base-deps.sh	21	# CUDA_MINOR_VERSION ("8" \| "0" \| "")	COMMENT
LOW	.docker/install-base-deps.sh	121	# above installs the Mesa drivers. Install it here so there is something to copy.	COMMENT
LOW	.docker/apt-mirror.sh	1	#!/bin/sh	COMMENT
LOW	core/trace/audio_snippet.go	21	// ~-90 dBFS least-significant-bit floor, so it reads unambiguously as	COMMENT
LOW	core/trace/backend_trace.go	261	func CancelBackendTrace(id string) {	COMMENT
LOW	core/config/model_config.go	101	// gRPCPredictOpts), so jinja-templated models that key on it — e.g. gpt-oss	COMMENT
LOW	core/config/model_config.go	181	// can have a stricter ceiling than local models.	COMMENT
LOW	core/config/model_config.go	201	// configuration error and disables the MITM listener until resolved.	COMMENT
LOW	core/config/model_config.go	221	// routing (OpenAI client → Anthropic upstream, etc.) at the cost	COMMENT
LOW	core/config/model_config.go	241	// allowed (no-auth upstreams).	COMMENT
LOW	core/config/model_config.go	261		COMMENT
LOW	core/config/model_config.go	301	// input.Model in-place, then the standard model-resolution path picks	COMMENT
LOW	core/config/model_config.go	321	// "small talk", ...). The Score classifier picks the subset of	COMMENT
LOW	core/config/model_config.go	341	// against (Arch-Router-1.5B is the canonical choice).	COMMENT
LOW	core/config/model_config.go	361	// schema). Override when the classifier model was trained on a	COMMENT
LOW	core/config/model_config.go	381	ScoreNormalization string `yaml:"score_normalization,omitempty" json:"score_normalization,omitempty"`	COMMENT
LOW	core/config/model_config.go	401	// cache. Pairs naturally with a larger / slower classifier model: the	COMMENT
LOW	core/config/model_config.go	421	// cached so they can't poison future paraphrases.	COMMENT
LOW	core/config/model_config.go	441	type RouterKNNConfig struct {	COMMENT
LOW	core/config/model_config.go	461	// similar than this to the probe cannot vote, and when none clear	COMMENT
LOW	core/config/model_config.go	541	// HasRouter returns true when the model declares a router config with	COMMENT
LOW	core/config/model_config.go	581		COMMENT
LOW	core/config/model_config.go	601	// Builtins names the built-in pattern groups this (pattern) detector	COMMENT
LOW	core/config/model_config.go	641	copy(out, c.PII.Detectors)	COMMENT
LOW	core/config/model_config.go	761	TTS string `yaml:"tts,omitempty" json:"tts,omitempty"`	COMMENT
LOW	core/config/model_config.go	781	// the pipeline's LLM without editing the LLM model config. Overrides the LLM's	COMMENT
LOW	core/config/model_config.go	801		COMMENT
LOW	core/config/model_config.go	821	// option's canned reply / tool call is emitted, so weak hardware	COMMENT
LOW	core/config/model_config.go	901	type PipelineCompaction struct {	COMMENT
LOW	core/config/model_config.go	1041	// trust the session).	COMMENT
LOW	core/config/model_config.go	1181		COMMENT
LOW	core/config/model_config.go	1481	threads := lo.threads	COMMENT
LOW	core/config/model_config.go	1941	// prompt cache with generation.	COMMENT
LOW	core/config/gguf.go	21	// (FLAG_CHAT, jinja templating): surfacing it in chat pickers defeats the	COMMENT
LOW	core/config/runtime_settings_startup.go	81	f.apply(o, settings)	COMMENT
LOW	core/config/vllm_spec.go	1	package config	COMMENT
LOW	core/config/vllm_spec.go	21	// speculative decoding can be auto-enabled.	COMMENT
LOW	core/config/vllm_spec.go	61	return len(c.DFlashConfig) > 0 \|\|	COMMENT
LOW	core/config/model_load_budget.go	1	package config	COMMENT
LOW	core/config/model_load_budget.go	21	// failure LATENCY on a load that was going to fail anyway, while erring	COMMENT
LOW	core/config/model_load_budget.go	41	)	COMMENT
LOW	core/config/runtime_settings_persist.go	41	// faithful partial update: a caller that submits only the field it owns	COMMENT
LOW	core/config/model_capabilities.go	41	}	COMMENT
LOW	core/config/runtime_settings_persist_test.go	141	got, err := cfg.ReadPersistedSettings()	COMMENT
LOW	core/config/defaults.go	1	package config	COMMENT
LOW	core/config/defaults.go	21	// DefaultAutoContextSize caps the context we auto-derive from a GGUF when the	COMMENT
LOW	core/config/context_fit.go	21	// candidate that fits.	COMMENT
LOW	core/config/context_fit.go	81	// not clamp the context.	COMMENT
1006 more matches not shown…

Cross-File Repetition118 hits · 590 pts

Severity	File	Line	Snippet	Context
HIGH	backend/python/pocket-tts/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/vibevoice/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/kitten-tts/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/whisperx/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/chatterbox/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/coqui/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/moonshine/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/neutts/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/qwen-tts/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/kokoro/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/faster-whisper/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/fish-speech/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/voxcpm/backend.py	0	backendservicer is the class that implements the grpc service	STRING
HIGH	backend/python/pocket-tts/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/vibevoice/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/kitten-tts/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/rerankers/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/diffusers/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/whisperx/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/chatterbox/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/coqui/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/moonshine/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/transformers/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/neutts/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/qwen-tts/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/kokoro/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/fish-speech/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/voxcpm/test.py	0	testbackendservicer is the class that tests the grpc service	STRING
HIGH	backend/python/pocket-tts/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/vibevoice/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/kitten-tts/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/rerankers/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/diffusers/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/whisperx/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/chatterbox/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/coqui/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/moonshine/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/transformers/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/neutts/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/qwen-tts/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/kokoro/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/fish-speech/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/voxcpm/test.py	0	this method sets up the grpc service by starting the server	STRING
HIGH	backend/python/pocket-tts/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/vibevoice/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/kitten-tts/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/rerankers/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/diffusers/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/whisperx/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/chatterbox/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/coqui/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/moonshine/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/transformers/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/neutts/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/qwen-tts/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/kokoro/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/fish-speech/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/voxcpm/test.py	0	this method tears down the grpc service by terminating the server	STRING
HIGH	backend/python/pocket-tts/test.py	0	this method tests if the server starts up successfully	STRING
HIGH	backend/python/vibevoice/test.py	0	this method tests if the server starts up successfully	STRING
58 more matches not shown…

Excessive Try-Catch Wrapping418 hits · 544 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_invariant_backend.py	37	except Exception:	CODE
MEDIUM	backend/python/liquid-audio/backend.py	147	def LoadModel(self, request, context):	CODE
MEDIUM	backend/python/liquid-audio/backend.py	218	def Predict(self, request, context):	CODE
MEDIUM	backend/python/liquid-audio/backend.py	229	def PredictStream(self, request, context):	CODE
MEDIUM	backend/python/liquid-audio/backend.py	301	def TTS(self, request, context):	CODE
MEDIUM	backend/python/liquid-audio/backend.py	517	def AudioTranscription(self, request, context):	CODE
MEDIUM	backend/python/liquid-audio/backend.py	618	def _run_training(self, request, job):	CODE
LOW⚡	backend/python/liquid-audio/backend.py	127	except Exception:	CODE
LOW⚡	backend/python/liquid-audio/backend.py	135	except Exception:	CODE
LOW⚡	backend/python/liquid-audio/backend.py	138	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	212	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	222	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	233	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	293	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	357	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	373	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	562	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	626	except Exception as exc:	CODE
LOW	backend/python/liquid-audio/backend.py	687	except Exception:	CODE
LOW	backend/python/pocket-tts/backend.py	117	except Exception as e:	CODE
LOW	backend/python/pocket-tts/backend.py	120	except Exception as err:	CODE
LOW	backend/python/pocket-tts/backend.py	150	except Exception as e:	CODE
MEDIUM	backend/python/pocket-tts/backend.py	151	print(f"Error loading voice from {voice_input}: {e}", file=sys.stderr)	CODE
LOW	backend/python/pocket-tts/backend.py	220	except Exception as err:	CODE
MEDIUM	backend/python/pocket-tts/backend.py	221	print(f"Error in TTS: {err}", file=sys.stderr)	CODE
MEDIUM	backend/python/pocket-tts/backend.py	154	def TTS(self, request, context):	CODE
LOW	backend/python/pocket-tts/test.py	43	except Exception as err:	CODE
LOW	backend/python/pocket-tts/test.py	61	except Exception as err:	CODE
LOW	backend/python/pocket-tts/test.py	98	except Exception as err:	CODE
LOW	backend/python/pocket-tts/test.py	137	except Exception as err:	CODE
LOW	backend/python/ace-step/backend.py	291	except Exception:	CODE
LOW	backend/python/ace-step/backend.py	369	except Exception as e:	CODE
LOW	backend/python/ace-step/backend.py	391	except Exception as err:	CODE
MEDIUM	backend/python/ace-step/backend.py	308	def LoadModel(self, request, context):	CODE
LOW	backend/python/vibevoice/backend.py	272	except Exception as e:	CODE
MEDIUM	backend/python/vibevoice/backend.py	276	print("Error loading the ASR model. Trying to use SDPA.", file=sys.stderr)	CODE
LOW	backend/python/vibevoice/backend.py	330	except Exception as e:	CODE
MEDIUM	backend/python/vibevoice/backend.py	334	print("Error loading the model. Trying to use SDPA. However, note that only flash_attention_2 ha	CODE
LOW	backend/python/vibevoice/backend.py	359	except Exception as err:	CODE
MEDIUM	backend/python/vibevoice/backend.py	463	print(f"Error accessing self.device: {e}, falling back to CPU", file=sys.stderr)	CODE
LOW	backend/python/vibevoice/backend.py	577	except Exception as err:	CODE
MEDIUM	backend/python/vibevoice/backend.py	578	print(f"Error in TTS: {err}", file=sys.stderr)	CODE
MEDIUM	backend/python/vibevoice/backend.py	598	print(f"Error: Audio file not found: {audio_path}", file=sys.stderr)	CODE
LOW	backend/python/vibevoice/backend.py	691	except Exception as e:	CODE
LOW	backend/python/vibevoice/backend.py	717	except Exception as err:	CODE
MEDIUM	backend/python/vibevoice/backend.py	718	print(f"Error in AudioTranscription: {err}", file=sys.stderr)	CODE
MEDIUM	backend/python/vibevoice/backend.py	475	def TTS(self, request, context):	CODE
LOW	backend/python/vibevoice/test.py	48	except Exception as err:	CODE
LOW	backend/python/vibevoice/test.py	66	except Exception as err:	CODE
LOW	backend/python/vibevoice/test.py	90	except Exception as err:	CODE
LOW	backend/python/vibevoice/test.py	128	except Exception as err:	CODE
LOW	backend/python/vibevoice/test.py	203	except Exception as err:	CODE
LOW	backend/python/kitten-tts/backend.py	56	except Exception as err:	CODE
LOW	backend/python/kitten-tts/backend.py	74	except Exception as err:	CODE
MEDIUM	backend/python/kitten-tts/backend.py	62	def TTS(self, request, context):	CODE
LOW	backend/python/kitten-tts/test.py	41	except Exception as err:	CODE
LOW	backend/python/kitten-tts/test.py	59	except Exception as err:	CODE
LOW	backend/python/kitten-tts/test.py	78	except Exception as err:	CODE
LOW	backend/python/rerankers/backend.py	69	except Exception as err:	CODE
LOW	backend/python/rerankers/test.py	41	except Exception as err:	CODE
358 more matches not shown…

Hyper-Verbose Identifiers204 hits · 235 pts

Severity	File	Line	Snippet	Context
LOW	core/http/react-ui/src/utils/artifacts.js	151	export function renderMarkdownWithArtifacts(text, messageIndex) {	CODE
LOW	core/http/react-ui/src/utils/cmYamlComplete.js	116	export function createYamlCompletionSource(fields) {	CODE
LOW	core/http/react-ui/src/utils/transferRate.js	3	export function createTransferRateSampler() {	CODE
LOW	core/http/react-ui/src/pages/Usage.jsx	182	function generateFutureBucketLabels(lastBucket, count, period) {	CODE
LOW	core/http/react-ui/src/pages/Middleware.jsx	419	function formatDecisionScoreSuffix(d, activeSet) {	CODE
LOW	core/http/react-ui/src/pages/InstalledModels.jsx	56	export function ModelLifecycleDetailShell({	CODE
LOW	core/http/react-ui/src/pages/BackendLogs.jsx	279	function DistributedBackendLogsResolver({ modelId, fromTimestamp }) {	CODE
LOW	core/http/static/talk.js	138	function updateStreamingTranscript(role, delta) {	CODE
LOW	core/http/static/talk.js	148	function finalizeStreamingTranscript(role, fullText) {	CODE
LOW	core/http/static/chat.js	544	function startTokensPerSecondInterval() {	CODE
LOW	core/http/static/chat.js	640	function stopTokensPerSecondInterval() {	CODE
LOW	core/http/static/chat.js	707	function updateMaxTokensPerSecondBadge(chatId, maxRate) {	CODE
LOW	core/http/static/chat.js	729	function scrollThinkingBoxToBottom() {	CODE
LOW	core/http/static/chat.js	1015	function processAndSendMessage(inputValue) {	CODE
LOW	core/http/static/assets/pdf.min.js	22	!function webpackUniversalModuleDefinition(t,e){"object"==typeof exports&&"object"==typeof module?module.exports=t.pdfjs	CODE
LOW	tests/test_invariant_backend.py	15	def test_dataset_source_path_traversal_blocked(dataset_source):	CODE
LOW	backend/python/liquid-audio/backend.py	54	def _patch_liquid_audio_local_paths():	CODE
LOW	backend/python/liquid-audio/backend.py	69	def _local_first_snapshot_download(repo_id, revision=None, **kwargs):	CODE
LOW	backend/python/liquid-audio/test.py	51	def test_load_finetune_mode_without_weights(self):	CODE
LOW	backend/python/pocket-tts/test.py	104	def test_tts_with_default_voice(self):	CODE
LOW	backend/python/ace-step/test.py	34	def test_sound_generation_minimal(self):	CODE
LOW	backend/python/tinygrad/test.py	63	def test_multi_call_and_thinking(self):	CODE
LOW	backend/python/tinygrad/test.py	76	def test_no_tool_call_is_passthrough(self):	CODE
LOW	backend/python/tinygrad/test.py	111	def test_keymap_renames_every_hf_key(self):	CODE
LOW⚡	backend/python/tinygrad/test.py	129	def test_tied_embedding_fallback_when_lm_head_missing(self):	CODE
LOW⚡	backend/python/tinygrad/test.py	135	def test_unknown_keys_are_skipped(self):	CODE
LOW⚡	backend/python/tinygrad/test.py	144	def test_qkv_bias_models_rejected(self):	CODE
LOW	backend/python/nemo/backend.py	116	def _build_segments_with_words(self, hypothesis, stride, timestamp_granularities=None):	CODE
LOW⚡	backend/python/diffusers/test.py	195	def test_resolve_pipeline_class_by_name(self):	CODE
LOW⚡	backend/python/diffusers/test.py	200	def test_resolve_pipeline_class_by_name_case_insensitive(self):	CODE
LOW⚡	backend/python/diffusers/test.py	206	def test_resolve_pipeline_class_by_task(self):	CODE
LOW⚡	backend/python/diffusers/test.py	216	def test_resolve_pipeline_class_unknown_name_raises(self):	CODE
LOW⚡	backend/python/diffusers/test.py	226	def test_resolve_pipeline_class_unknown_task_raises(self):	CODE
LOW⚡	backend/python/diffusers/test.py	236	def test_resolve_pipeline_class_no_params_raises(self):	CODE
LOW⚡	backend/python/diffusers/test.py	270	def test_discover_diffusers_classes_caches_results(self):	CODE
LOW⚡	backend/python/diffusers/test.py	278	def test_discover_diffusers_classes_exclude_base(self):	CODE
LOW⚡	backend/python/diffusers/test.py	285	def test_get_available_classes(self):	CODE
LOW	backend/python/diffusers/test.py	108	def test_camel_to_kebab_conversion(self):	CODE
LOW	backend/python/diffusers/test.py	121	def test_extract_task_keywords(self):	CODE
LOW	backend/python/diffusers/test.py	141	def test_discover_pipelines_finds_known_classes(self):	CODE
LOW	backend/python/diffusers/test.py	162	def test_discover_pipelines_caches_results(self):	CODE
LOW	backend/python/diffusers/test.py	171	def test_get_available_pipelines(self):	CODE
LOW	backend/python/diffusers/test.py	252	def test_get_pipeline_info_unknown_raises(self):	CODE
LOW	backend/python/diffusers/test.py	259	def test_discover_diffusers_classes_pipelines(self):	CODE
LOW	backend/python/diffusers/test.py	300	def test_load_pipeline_requires_model_id(self):	CODE
LOW	backend/python/diffusers/test.py	307	def test_resolve_with_model_id_uses_diffusion_pipeline_fallback(self):	CODE
LOW	backend/python/diffusers/test.py	326	def test_options_merged_into_pipeline_kwargs(self):	CODE
LOW	backend/python/diffusers/diffusers_dynamic_loader.py	124	def discover_diffusers_classes(	CODE
LOW	backend/python/chatterbox/backend.py	54	def split_text_at_word_boundary(text, max_length=250):	CODE
LOW	backend/python/vllm/test.py	171	def test_apply_engine_args_known_keys(self):	CODE
LOW	backend/python/vllm/test.py	192	def test_apply_engine_args_unknown_key_raises(self):	CODE
LOW	backend/python/vllm/test.py	209	def test_apply_engine_args_empty_passthrough(self):	CODE
LOW	backend/python/vllm/test.py	335	def test_buffer_path_no_markup_no_duplication(self):	CODE
LOW	backend/python/vllm/test.py	388	def test_native_streaming_progressive_plain_text(self):	CODE
LOW	backend/python/vllm/test.py	403	def extract_tool_calls_streaming(	CODE
LOW	backend/python/vllm/test.py	440	def test_native_streaming_tool_call_no_markup_leak(self):	CODE
LOW	backend/python/vllm/test.py	454	def extract_tool_calls_streaming(	CODE
LOW	backend/python/vllm/test.py	489	def test_native_streaming_parser_exception_falls_back_to_buffer(self):	CODE
LOW	backend/python/vllm/test.py	499	def extract_tool_calls_streaming(self, a, *kw):	CODE
LOW	backend/python/vllm/test.py	523	def test_no_tool_parser_unchanged_per_delta_stream(self):	CODE
144 more matches not shown…

Decorative Section Separators44 hits · 146 pts

Severity	File	Line	Snippet	Context
MEDIUM	core/schema/localai.go	271	// ─── Face recognition ──────────────────────────────────────────────	COMMENT
MEDIUM	core/schema/localai.go	398	// ─── Voice (speaker) recognition ───────────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/Quantize.jsx	19	// ── Reusable sub-components ──────────────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/Quantize.jsx	191	// ── Main page ────────────────────────────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/FaceRecognition.jsx	84	// ──────────────────────────── Analyze ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/FaceRecognition.jsx	218	// ──────────────────────────── Compare ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/FaceRecognition.jsx	333	// ──────────────────────────── Enrollment (register / identify / forget) ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/FaceRecognition.jsx	516	// ──────────────────────────── Embedding ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/FaceRecognition.jsx	570	// ──────────────────────────── Small shared bits ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/VoiceRecognition.jsx	86	// ──────────────────────────── Analyze ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/VoiceRecognition.jsx	196	// ──────────────────────────── Compare ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/VoiceRecognition.jsx	288	// ──────────────────────────── Enrollment ────────────────────────────	COMMENT
MEDIUM	core/http/react-ui/src/pages/VoiceRecognition.jsx	472	// ──────────────────────────── Embedding ────────────────────────────	COMMENT
MEDIUM	core/services/facerecognition/registry_test.go	134	// ─── fake gRPC backend ───────────────────────────────────────────────	COMMENT
MEDIUM	tests/e2e-backends/backend_test.go	794	// ─── face recognition specs ─────────────────────────────────────────	COMMENT
MEDIUM	tests/e2e-backends/backend_test.go	983	// ─── voice (speaker) recognition specs ──────────────────────────────	COMMENT
MEDIUM	tests/e2e-backends/backend_test.go	1231	// ─── ds4 layer-split distributed inference (opt-in, hardware-gated) ──────────	COMMENT
MEDIUM	backend/go/parakeet-cpp/boundary.go	23	// open ───────────► closed	COMMENT
MEDIUM	backend/go/parakeet-cpp/boundary.go	27	// └──────────────────┘	COMMENT
MEDIUM	backend/go/vllm-cpp/govllmcpp.go	133	// ── Video+audio generation (ABI v12, MiniMax-H3) ────────────────────────────	COMMENT
MEDIUM⚡	backend/python/tinygrad/backend.py	57	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/tinygrad/backend.py	64	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/tinygrad/backend.py	72	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/tinygrad/backend.py	74	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/tinygrad/backend.py	158	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/tinygrad/backend.py	160	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/diffusers/backend.py	245	# ================================================================	COMMENT
MEDIUM	backend/python/diffusers/backend.py	249	# ================================================================	COMMENT
MEDIUM	backend/python/diffusers/backend.py	460	# ================================================================	COMMENT
MEDIUM	backend/python/diffusers/backend.py	463	# ================================================================	COMMENT
MEDIUM	backend/python/common/libbackend.sh	34	# =====================================================================	COMMENT
MEDIUM⚡	backend/python/trl/reward_functions.py	28	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/trl/reward_functions.py	30	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/trl/reward_functions.py	116	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/trl/reward_functions.py	118	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/trl/reward_functions.py	130	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/trl/reward_functions.py	132	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/trl/reward_functions.py	192	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	backend/python/trl/reward_functions.py	194	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	backend/python/insightface/engines.py	61	# ─── Antispoofer (Silent-Face MiniFASNet) ──────────────────────────────	COMMENT
MEDIUM	backend/python/insightface/engines.py	174	# ─── InsightFaceEngine ────────────────────────────────────────────────	COMMENT
MEDIUM	backend/python/insightface/engines.py	375	# ─── OnnxDirectEngine ─────────────────────────────────────────────────	COMMENT
MEDIUM	backend/python/insightface/engines.py	481	# ─── helpers ──────────────────────────────────────────────────────────	COMMENT
MEDIUM⚡	.github/backend-matrix.yml	7	# --------------------------------------------------------	COMMENT

Self-Referential Comments39 hits · 122 pts

Severity	File	Line	Snippet	Context
MEDIUM	.devcontainer-scripts/utils.sh	3	# This file contains some really simple functions that are useful when building up customization scripts.	COMMENT
MEDIUM	backend/go/sam3-cpp/test.sh	22	# Create a test image (4x4 red pixel PNG) using base64	COMMENT
MEDIUM	backend/python/pocket-tts/backend.py	241	# Define the signal handler function	COMMENT
MEDIUM	backend/python/vibevoice/backend.py	739	# Define the signal handler function	COMMENT
MEDIUM	backend/python/vibevoice/test.py	102	# Create a temporary directory for the output audio file	COMMENT
MEDIUM	backend/python/vibevoice/test.py	142	# Create a temporary directory for the audio file	COMMENT
MEDIUM	backend/python/kitten-tts/backend.py	92	# Define the signal handler function	COMMENT
MEDIUM	backend/python/rerankers/backend.py	112	# Define the signal handler function	COMMENT
MEDIUM⚡	backend/python/nemo/backend.py	293	# Create a single segment with the full transcription	COMMENT
MEDIUM	backend/python/diffusers/backend.py	1114	# Define the signal handler function	COMMENT
MEDIUM	backend/python/diffusers/test.py	343	# Create a tiny source image for the request's src field	COMMENT
MEDIUM	backend/python/whisperx/backend.py	153	# Define the signal handler function	COMMENT
MEDIUM	backend/python/whisperx/test.py	71	# Create a temporary directory for the audio file	COMMENT
MEDIUM	backend/python/chatterbox/backend.py	266	# Define the signal handler function	COMMENT
MEDIUM	backend/python/coqui/backend.py	113	# Define the signal handler function	COMMENT
MEDIUM	backend/python/moonshine/backend.py	143	# Define the signal handler function	COMMENT
MEDIUM	backend/python/moonshine/test.py	71	# Create a temporary directory for the audio file	COMMENT
MEDIUM	backend/python/common/libbackend.sh	338	# This function is idempotent, so you can call it as many times as you want and it will	COMMENT
MEDIUM	backend/python/neutts/backend.py	154	# Define the signal handler function	COMMENT
MEDIUM	backend/python/qwen-tts/backend.py	966	# Define the signal handler function	COMMENT
MEDIUM	backend/python/rfdetr/backend.py	154	# Define the signal handler function	COMMENT
MEDIUM	backend/python/kokoro/backend.py	99	# Define the signal handler function	COMMENT
MEDIUM	backend/python/faster-whisper/backend.py	101	# Define the signal handler function	COMMENT
MEDIUM	backend/python/fish-speech/backend.py	451	# Define the signal handler function	COMMENT
MEDIUM	backend/python/voxcpm/backend.py	316	# Define the signal handler function	COMMENT
MEDIUM	scripts/nats-auth-setup.sh	48	# Create the operator if it does not exist, otherwise select it.	COMMENT
MEDIUM	scripts/nats-auth-setup.sh	70	# Create the LocalAI account if it does not exist.	COMMENT
MEDIUM	scripts/nats-auth-setup.sh	79	# Create the frontend service user if it does not exist.	COMMENT
MEDIUM⚡	.github/backend-matrix.yml	4	# This file is NOT a workflow — it has no top-level 'on:' or 'jobs:'.	COMMENT
MEDIUM⚡	.github/backend-matrix.yml	8	# This file is the source of truth for which OS each backend is built and	COMMENT
MEDIUM	.github/workflows/gallery-agent.yaml	162	# Create a formatted list of added models with URLs	COMMENT
MEDIUM	.github/workflows/notify-releases.yaml	25	# Define the LocalAI API endpoint	COMMENT
MEDIUM	.github/workflows/notify-releases.yaml	28	# Create a JSON payload using jq to handle special characters	COMMENT
MEDIUM	.github/workflows/disabled/notify-models.yaml	39	# Define the LocalAI API endpoint	COMMENT
MEDIUM	.github/workflows/disabled/notify-models.yaml	42	# Create a JSON payload using jq to handle special characters	COMMENT
MEDIUM	.github/workflows/disabled/notify-models.yaml	121	# Define the LocalAI API endpoint	COMMENT
MEDIUM	.github/workflows/disabled/notify-models.yaml	124	# Create a JSON payload using jq to handle special characters	COMMENT
MEDIUM	.github/workflows/disabled/comment-pr.yaml	39	# Define the LocalAI API endpoint	COMMENT
MEDIUM	.github/workflows/disabled/comment-pr.yaml	42	# Create a JSON payload using jq to handle special characters	COMMENT

Fake / Example Data77 hits · 84 pts

Severity	File	Line	Snippet	Context
LOW⚡	core/http/auth/branding_routes_test.go	75	user := createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/branding_routes_test.go	84	admin := createTestUser(db, "admin@example.com", auth.RoleAdmin, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/branding_routes_test.go	93	admin := createTestUser(db, "admin@example.com", auth.RoleAdmin, auth.ProviderGitHub)	CODE
LOW	core/http/auth/middleware_test.go	214	user = createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW	core/http/auth/middleware_test.go	484	admin := createTestUser(db, "admin@example.com", auth.RoleAdmin, auth.ProviderGitHub)	CODE
LOW	core/http/auth/middleware_test.go	493	user := createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/middleware_test.go	509	admin := createTestUser(db, "admin@example.com", auth.RoleAdmin, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/middleware_test.go	518	user := createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/middleware_test.go	527	user := createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/roles_test.go	35	role := auth.AssignRole(db, "admin@example.com", "admin@example.com")	CODE
LOW⚡	core/http/auth/roles_test.go	42	role := auth.AssignRole(db, "Admin@Example.COM", "admin@example.com")	CODE
LOW⚡	core/http/auth/roles_test.go	49	role := auth.AssignRole(db, "other@example.com", "admin@example.com")	CODE
LOW⚡	core/http/auth/roles_test.go	69	user := createTestUser(db, "user@example.com", auth.RoleUser, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/roles_test.go	71	promoted := auth.MaybePromote(db, user, "admin@example.com")	CODE
LOW⚡	core/http/auth/roles_test.go	77	user := createTestUser(db, "admin@example.com", auth.RoleAdmin, auth.ProviderGitHub)	CODE
LOW⚡	core/http/auth/oauth_email_decision_test.go	13	Expect(emailForRoleDecision("admin@example.com", true)).	CODE
LOW⚡	core/http/auth/oauth_email_decision_test.go	14	To(Equal("admin@example.com"))	CODE
LOW⚡	core/http/auth/oauth_email_decision_test.go	18	Expect(emailForRoleDecision("admin@example.com", false)).	CODE
LOW	core/http/auth/oauth_email_decision_test.go	46	role := AssignRole(db, emailForRoleDecision("admin@example.com", false), "admin@example.com")	CODE
LOW	core/http/auth/oauth_email_decision_test.go	52	role := AssignRole(db, emailForRoleDecision("admin@example.com", true), "admin@example.com")	CODE
LOW	core/http/react-ui/public/locales/it/chat.json	91	"placeholder": "Messaggio...",	CODE
LOW	core/http/react-ui/public/locales/it/models.json	104	"placeholder": "Cerca modelli...",	CODE
LOW	core/http/react-ui/public/locales/it/agents.json	23	"placeholder": "Cerca agenti...",	CODE
LOW	core/http/react-ui/public/locales/it/home.json	31	"placeholder": "Messaggio...",	CODE
LOW	core/http/react-ui/public/locales/it/skills.json	25	"placeholder": "Cerca competenze..."	CODE
LOW	core/http/react-ui/public/locales/zh-CN/chat.json	91	"placeholder": "消息...",	CODE
LOW	core/http/react-ui/public/locales/zh-CN/models.json	104	"placeholder": "搜索模型...",	CODE
LOW	core/http/react-ui/public/locales/zh-CN/agents.json	23	"placeholder": "搜索智能体...",	CODE
LOW	core/http/react-ui/public/locales/zh-CN/home.json	31	"placeholder": "消息...",	CODE
LOW	core/http/react-ui/public/locales/zh-CN/skills.json	25	"placeholder": "搜索技能..."	CODE
LOW	core/http/react-ui/public/locales/pt-BR/chat.json	92	"placeholder": "Mensagem...",	CODE
LOW	…re/http/react-ui/public/locales/pt-BR/modelEditor.json	29	"placeholder": "meu-nome-de-modelo",	CODE
LOW	core/http/react-ui/public/locales/pt-BR/models.json	114	"placeholder": "Buscar modelos...",	CODE
LOW	core/http/react-ui/public/locales/pt-BR/agents.json	23	"placeholder": "Buscar agentes...",	CODE
LOW	core/http/react-ui/public/locales/pt-BR/media.json	200	"placeholder": "Buscar vozes, idiomas ou transcrições"	CODE
LOW	core/http/react-ui/public/locales/pt-BR/home.json	31	"placeholder": "Mensagem...",	CODE
LOW	core/http/react-ui/public/locales/pt-BR/skills.json	25	"placeholder": "Buscar habilidades..."	CODE
LOW	core/http/react-ui/public/locales/de/chat.json	91	"placeholder": "Nachricht...",	CODE
LOW	core/http/react-ui/public/locales/de/models.json	104	"placeholder": "Modelle suchen...",	CODE
LOW	core/http/react-ui/public/locales/de/agents.json	23	"placeholder": "Agenten suchen...",	CODE
LOW	core/http/react-ui/public/locales/de/home.json	31	"placeholder": "Nachricht...",	CODE
LOW	core/http/react-ui/public/locales/de/skills.json	25	"placeholder": "Fähigkeiten suchen..."	CODE
LOW	core/http/react-ui/public/locales/ko/chat.json	91	"placeholder": "메시지...",	CODE
LOW	core/http/react-ui/public/locales/ko/models.json	110	"placeholder": "모델 검색...",	CODE
LOW	core/http/react-ui/public/locales/ko/agents.json	23	"placeholder": "에이전트 검색...",	CODE
LOW	core/http/react-ui/public/locales/ko/home.json	31	"placeholder": "메시지...",	CODE
LOW	core/http/react-ui/public/locales/ko/skills.json	25	"placeholder": "스킬 검색..."	CODE
LOW	core/http/react-ui/public/locales/id/chat.json	91	"placeholder": "Pesan...",	CODE
LOW	core/http/react-ui/public/locales/id/modelEditor.json	29	"placeholder": "nama-model-saya",	CODE
LOW	core/http/react-ui/public/locales/id/models.json	111	"placeholder": "Cari model...",	CODE
LOW	core/http/react-ui/public/locales/id/agents.json	23	"placeholder": "Cari agen...",	CODE
LOW	core/http/react-ui/public/locales/id/home.json	31	"placeholder": "Pesan...",	CODE
LOW	core/http/react-ui/public/locales/id/skills.json	25	"placeholder": "Cari skill..."	CODE
LOW	core/http/react-ui/public/locales/es/chat.json	91	"placeholder": "Mensaje...",	CODE
LOW	core/http/react-ui/public/locales/es/models.json	104	"placeholder": "Buscar modelos...",	CODE
LOW	core/http/react-ui/public/locales/es/agents.json	23	"placeholder": "Buscar agentes...",	CODE
LOW	core/http/react-ui/public/locales/es/home.json	31	"placeholder": "Mensaje...",	CODE
LOW	core/http/react-ui/public/locales/es/skills.json	25	"placeholder": "Buscar habilidades..."	CODE
LOW	core/http/react-ui/public/locales/en/chat.json	102	"placeholder": "Message...",	CODE
LOW	core/http/react-ui/public/locales/en/modelEditor.json	29	"placeholder": "my-model-name",	CODE
17 more matches not shown…

Deep Nesting76 hits · 62 pts

Severity	File	Line	Context
LOW	tests/test_invariant_backend.py	15	CODE
LOW	backend/python/liquid-audio/backend.py	119	CODE
LOW	backend/python/liquid-audio/backend.py	240	CODE
LOW	backend/python/liquid-audio/backend.py	381	CODE
LOW	backend/python/liquid-audio/backend.py	517	CODE
LOW	backend/python/pocket-tts/backend.py	54	CODE
LOW	backend/python/pocket-tts/backend.py	154	CODE
LOW	backend/python/ace-step/backend.py	82	CODE
LOW	backend/python/ace-step/backend.py	102	CODE
LOW	backend/python/ace-step/backend.py	308	CODE
LOW	backend/python/vibevoice/backend.py	56	CODE
LOW	backend/python/vibevoice/backend.py	475	CODE
LOW	backend/python/vibevoice/backend.py	584	CODE
LOW	backend/python/nemo/backend.py	48	CODE
LOW	backend/python/nemo/backend.py	227	CODE
LOW	backend/python/diffusers/backend.py	171	CODE
LOW	backend/python/diffusers/backend.py	509	CODE
LOW	backend/python/diffusers/backend.py	673	CODE
LOW	backend/python/diffusers/backend.py	914	CODE
LOW	backend/python/diffusers/diffusers_dynamic_loader.py	124	CODE
LOW	backend/python/diffusers/diffusers_dynamic_loader.py	313	CODE
LOW	backend/python/mlx-audio/backend.py	67	CODE
LOW	backend/python/mlx-audio/backend.py	303	CODE
LOW	backend/python/mlx-audio/backend.py	382	CODE
LOW	backend/python/llama-cpp-quantization/backend.py	362	CODE
LOW	backend/python/chatterbox/backend.py	135	CODE
LOW	backend/python/chatterbox/backend.py	196	CODE
LOW	backend/python/vllm/backend.py	149	CODE
LOW	backend/python/vllm/backend.py	394	CODE
LOW	backend/python/vllm/backend.py	526	CODE
LOW	backend/python/speaker-recognition/engines.py	117	CODE
LOW	backend/python/common/model_identity.py	226	CODE
LOW	backend/python/common/python_utils.py	10	CODE
LOW	backend/python/common/python_utils.py	40	CODE
LOW	backend/python/transformers/backend.py	64	CODE
LOW	backend/python/transformers/backend.py	491	CODE
LOW	backend/python/transformers/backend.py	557	CODE
LOW	backend/python/vllm-omni/backend.py	74	CODE
LOW	backend/python/vllm-omni/backend.py	355	CODE
LOW	backend/python/vllm-omni/backend.py	479	CODE
LOW	backend/python/vllm-omni/backend.py	663	CODE
LOW	backend/python/mlx-distributed/backend.py	619	CODE
LOW	backend/python/mlx-distributed/backend.py	399	CODE
LOW	backend/python/mlx-distributed/mlx_cache.py	76	CODE
LOW	backend/python/mlx-distributed/mlx_cache.py	169	CODE
LOW	backend/python/neutts/backend.py	50	CODE
LOW	backend/python/mlx-vlm/backend.py	58	CODE
LOW	backend/python/mlx-vlm/backend.py	262	CODE
LOW	backend/python/trl/backend.py	248	CODE
LOW	backend/python/trl/backend.py	644	CODE
LOW	backend/python/trl/backend.py	691	CODE
LOW	backend/python/trl/reward_functions.py	196	CODE
LOW	backend/python/sglang/backend.py	139	CODE
LOW	backend/python/sglang/backend.py	403	CODE
LOW	backend/python/qwen-asr/backend.py	97	CODE
LOW	backend/python/qwen-asr/backend.py	212	CODE
LOW	backend/python/qwen-tts/backend.py	86	CODE
LOW	backend/python/qwen-tts/backend.py	482	CODE
LOW	backend/python/qwen-tts/backend.py	690	CODE
LOW	backend/python/mlx/backend.py	252	CODE
16 more matches not shown…

Modern AI Meta-Vocabulary19 hits · 53 pts

Severity	File	Line	Snippet	Context
MEDIUM	README.md	190	- October 2025: [Model Context Protocol (MCP)](https://localai.io/docs/features/mcp/) support for agentic capabiliti	CODE
MEDIUM	core/config/model_config.go	267	// agentic/multi-turn workloads that re-send a large stable prefix.	COMMENT
MEDIUM	core/http/endpoints/openai/realtime.go	2139	// maxAssistantToolTurns caps the server-side agentic loop. Mirrors the	COMMENT
MEDIUM	core/http/endpoints/openai/realtime.go	2156	// the whole agentic tool-turn recursion: a single id, the output items as they	COMMENT
MEDIUM	core/http/endpoints/openai/realtime.go	2597	// triggerResponse; every turn (including agentic recursion) shares this id.	COMMENT
MEDIUM	core/http/endpoints/openai/realtime.go	2886	// whole turn (including the agentic recursion below) completes.	COMMENT
MEDIUM	core/http/endpoints/openai/realtime.go	2889	// so the model can speak the result. Mirrors the chat-side agentic loop	COMMENT
MEDIUM	core/http/endpoints/openai/inference_test.go	411	// Truncation labeling for agentic clients (cogito): a reasoning model	COMMENT
MEDIUM	core/http/react-ui/src/hooks/useChat.js	555	// Regular SSE streaming with client-side agentic loop support	COMMENT
MEDIUM	core/http/react-ui/src/hooks/useChat.js	572	let fullToolCalls = [] // Tool calls with id for agentic loop	CODE
MEDIUM	core/http/react-ui/src/hooks/useChat.js	727	// Client-side agentic loop: check for client tool calls	COMMENT
MEDIUM	core/http/react-ui/src/pages/Talk.jsx	211	// Mirrors the chat-page agentic loop: collect args from the model's	COMMENT
MEDIUM	core/services/agentpool/agent_pool.go	62	// to provide agentic capabilities integrated directly into LocalAI.	COMMENT
MEDIUM	website/content/blog/localai-since-march-2023.md	41	[LocalAI 4.0.0](https://github.com/mudler/LocalAI/releases/tag/v4.0.0) added native agentic orchestration with the [Agen	CODE
MEDIUM	tests/e2e-aio/models/embeddings.yaml	7	# sized up to the context window (core/backend/options.go EffectiveBatchSize) so	COMMENT
MEDIUM	backend/go/cloud-proxy/provider_anthropic.go	66	// calls that share it — the win on agentic/multi-turn workloads.	COMMENT
MEDIUM	backend/go/cloud-proxy/provider_anthropic.go	239	// cache-read rate on the next agentic iteration (max 4 breakpoints; we	COMMENT
MEDIUM	backend/python/tinygrad/backend.py	629	# Clamp to context window — truncate long inputs rather than blow up.	COMMENT
MEDIUM	docs/content/operations/middleware.md	691	embedding_model: nomic-embed-text-v1.5 # any loaded embedding model	CODE

Unused Imports60 hits · 52 pts

Severity	File	Line	Context
LOW	backend/python/vibevoice/backend.py	13	CODE
LOW	backend/python/kitten-tts/backend.py	14	CODE
LOW	backend/python/tinygrad/backend.py	28	CODE
LOW	backend/python/tinygrad/backend.py	52	CODE
LOW	backend/python/tinygrad/tool_parsers/mistral.py	11	CODE
LOW	backend/python/tinygrad/tool_parsers/__init__.py	7	CODE
LOW	backend/python/tinygrad/tool_parsers/__init__.py	9	CODE
LOW	backend/python/tinygrad/tool_parsers/__init__.py	9	CODE
LOW	backend/python/tinygrad/tool_parsers/__init__.py	9	CODE
LOW	backend/python/tinygrad/tool_parsers/hermes.py	17	CODE
LOW	backend/python/tinygrad/tool_parsers/llama3_json.py	15	CODE
LOW	backend/python/tinygrad/tool_parsers/llama3_json.py	19	CODE
LOW	backend/python/tinygrad/tool_parsers/qwen3_xml.py	21	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	2	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	4	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	73	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	73	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	73	CODE
LOW	backend/python/tinygrad/tool_parsers/base.py	73	CODE
LOW	backend/python/diffusers/backend.py	32	CODE
LOW	backend/python/diffusers/backend.py	32	CODE
LOW	backend/python/diffusers/backend.py	40	CODE
LOW	backend/python/diffusers/test.py	7	CODE
LOW	backend/python/diffusers/diffusers_dynamic_loader.py	33	CODE
LOW	backend/python/mlx-audio/backend.py	8	CODE
LOW	backend/python/mlx-audio/backend.py	9	CODE
LOW	backend/python/mlx-audio/backend.py	10	CODE
LOW	backend/python/mlx-audio/backend.py	11	CODE
LOW	backend/python/mlx-audio/backend.py	12	CODE
LOW	backend/python/vllm/backend.py	14	CODE
LOW	backend/python/vllm/backend.py	35	CODE
LOW	backend/python/speaker-recognition/backend.py	9	CODE
LOW	backend/python/speaker-recognition/engines.py	20	CODE
LOW	backend/python/speaker-recognition/test.py	10	CODE
LOW	backend/python/common/vllm_utils.py	12	CODE
LOW	backend/python/common/vllm_utils.py	12	CODE
LOW	backend/python/common/template/backend.py	2	CODE
LOW	backend/python/common/template/backend.py	3	CODE
LOW	backend/python/common/template/backend.py	4	CODE
LOW	backend/python/transformers/backend.py	14	CODE
LOW	backend/python/mlx-distributed/coordinator.py	7	CODE
LOW	backend/python/mlx-distributed/coordinator.py	8	CODE
LOW	backend/python/mlx-distributed/backend.py	23	CODE
LOW	backend/python/mlx-vlm/backend.py	11	CODE
LOW	backend/python/trl/backend.py	251	CODE
LOW	backend/python/sglang/backend.py	30	CODE
LOW	backend/python/sglang/backend.py	31	CODE
LOW	backend/python/sglang/backend.py	32	CODE
LOW	backend/python/sglang/backend.py	34	CODE
LOW	backend/python/qwen-tts/backend.py	12	CODE
LOW	backend/python/qwen-tts/backend.py	14	CODE
LOW	backend/python/qwen-tts/test.py	10	CODE
LOW	backend/python/rfdetr/backend.py	21	CODE
LOW	backend/python/rfdetr/backend.py	23	CODE
LOW	backend/python/insightface/backend.py	15	CODE
LOW	backend/python/insightface/engines.py	19	CODE
LOW	backend/python/insightface/test.py	10	CODE
LOW	backend/python/insightface/smoke.py	17	CODE
LOW	backend/python/mlx/test_mlx_cache.py	17	CODE
LOW	backend/python/faster-qwen3-tts/test.py	8	CODE

Structural Annotation Overuse26 hits · 46 pts

Severity	File	Line	Snippet	Context
LOW	core/gallery/upgrade.go	296	// Step 1: Download the new backend into the tmp directory	COMMENT
LOW⚡	core/gallery/upgrade.go	314	// Step 2: Validate — check that run.sh exists in the new content	COMMENT
LOW⚡	core/gallery/upgrade.go	321	// Step 3: Atomic swap — rename current to backup, then tmp to current	COMMENT
LOW	core/gallery/upgrade.go	337	// Step 4: Write updated metadata, preserving alias from old metadata	COMMENT
LOW⚡	core/gallery/upgrade.go	370	// Step 5: Re-register backends so the model loader picks up any changes	COMMENT
LOW⚡	core/gallery/upgrade.go	375	// Step 6: Remove backup	COMMENT
LOW	core/services/nodes/router_test.go	978	// Step 1: cached model found on old node	COMMENT
LOW⚡	core/services/nodes/router.go	663	// Step 1: Find and atomically lock a node with this model loaded	COMMENT
LOW⚡	core/services/nodes/router.go	668	// Step 2: model not loaded — it has to be cold-loaded.	COMMENT
LOW	tests/e2e/mock_backend_test.go	445	// Step 1: tokenize	COMMENT
LOW	tests/e2e/mock_backend_test.go	465	// Step 2: detokenize the returned token IDs	COMMENT
LOW	.agents/api-endpoints-and-auth.md	19	### Step 1: Create the handler	COMMENT
LOW	.agents/api-endpoints-and-auth.md	36	### Step 2: Register routes	COMMENT
LOW	.agents/api-endpoints-and-auth.md	48	### Step 3: Apply the right middleware	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	101	# Step 1: Download / resolve model	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	112	# Step 2: Convert to f16 GGUF	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	123	# Step 3: Quantize	COMMENT
LOW	backend/python/trl/backend.py	749	# Step 1: Merge LoRA into base model	COMMENT
LOW	backend/python/trl/backend.py	798	# Step 2: Convert to GGUF using convert_hf_to_gguf.py	COMMENT
LOW	docs/content/getting-started/models.md	185	### Step 1: Download a Model	COMMENT
LOW	docs/content/getting-started/models.md	200	### Step 2: Create a Configuration File (Optional)	COMMENT
LOW	docs/content/getting-started/models.md	217	### Step 3: Run LocalAI	COMMENT
LOW⚡	pkg/functions/parse_test.go	191	// Step 1: Replace single quotes around keys and values with double quotes	COMMENT
LOW⚡	pkg/functions/parse_test.go	193	// Step 2: Replace double quotes inside values with placeholders	COMMENT
LOW⚡	pkg/functions/parse_test.go	224	// Step 1: Replace single quotes around keys and values with double quotes	COMMENT
LOW⚡	pkg/functions/parse_test.go	226	// Step 2: Replace double quotes inside values with placeholders	COMMENT

Redundant / Tautological Comments27 hits · 43 pts

Severity	File	Line	Snippet	Context
LOW	backend/python/vibevoice/backend.py	104	# Check if ASR mode is enabled	COMMENT
LOW	backend/python/vibevoice/backend.py	194	# Check if the directory exists	COMMENT
LOW	backend/python/vibevoice/test.py	15	# Check if we should skip ASR tests (they require large models ~14B parameters total)	COMMENT
LOW	backend/python/vibevoice/test.py	188	# Check if segments is iterable (has length)	COMMENT
LOW	backend/python/diffusers/backend.py	357	# Check if loading from single file (GGUF)	COMMENT
LOW	backend/python/diffusers/backend.py	411	# Check if loading from single file (GGUF)	COMMENT
LOW	backend/python/diffusers/backend.py	638	# Check if its a local file and not a directory ( we load lora differently for a safetensor file )	COMMENT
LOW	backend/python/diffusers/diffusers_dynamic_loader.py	485	# Check if the class has from_single_file method	COMMENT
LOW⚡	backend/python/mlx-audio/backend.py	264	# Check if model supports language codes (primarily Kokoro)	COMMENT
LOW	backend/python/mlx-audio/backend.py	133	# Check if model is loaded	COMMENT
LOW	backend/python/mlx-audio/backend.py	156	# Check if model supports language codes (primarily Kokoro)	COMMENT
LOW	backend/python/mlx-audio/backend.py	235	# Check if model is loaded	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	377	# Check if the thread is still alive	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	384	# Check if client disconnected	COMMENT
LOW	backend/python/chatterbox/backend.py	67	# Check if adding this word would exceed the limit	COMMENT
LOW	backend/python/chatterbox/backend.py	220	# Check if text exceeds 250 characters	COMMENT
LOW	backend/python/vllm/backend.py	334	# Check if we have one result at least	COMMENT
LOW	backend/python/moonshine/test.py	111	# Check if segments is iterable (has length)	COMMENT
LOW⚡	backend/python/qwen-tts/backend.py	476	# Check if it looks like a file path (contains / or \ and ends with common text file extensions)	COMMENT
LOW	backend/python/qwen-tts/backend.py	287	# Check if it's a missing feature extractor/tokenizer error	COMMENT
LOW	backend/python/qwen-tts/backend.py	506	# Check if file exists and is readable	COMMENT
LOW	backend/python/qwen-tts/backend.py	692	# Check if dst is provided	COMMENT
LOW	backend/python/qwen-tts/backend.py	753	# Check if multi-voice mode is active (voices dict is populated)	COMMENT
LOW	backend/python/qwen-tts/backend.py	821	# Check if we should use cached prompt	COMMENT
LOW	backend/python/fish-speech/test.py	95	# Check if process died	COMMENT
LOW	.github/check_and_update.py	32	# Check if 'hasUnsafeFile' exists in the response	COMMENT
LOW⚡	.github/check_and_update.py	48	# Check if the repo is flagged as dangerous by HF	COMMENT

AI Slop Vocabulary15 hits · 42 pts

Severity	File	Line	Snippet	Context
MEDIUM	core/gallery/importers/bark.go	25	// DiscoverModelConfig leaves HuggingFace nil. To keep detection robust,	COMMENT
MEDIUM	core/http/react-ui/src/pages/AgentChat.jsx	116	// Maps backend messageID → conversationId for robust SSE routing across navigations.	COMMENT
MEDIUM	tests/e2e/distributed/prefix_cache_routing_test.go	182	// provider decision directly is the robust check.	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	203	# If the model_path is already a GGUF file, just use it as-is	COMMENT
MEDIUM	backend/python/speaker-recognition/engines.py	71	# checkpoint (Audeering wav2vec2-large-robust-24-ft-age-gender) uses a	COMMENT
MEDIUM	backend/cpp/privacy-filter/package.sh	17	# grpc++/protobuf/absl stack) by walking ldd — robust to whichever of those are	COMMENT
MEDIUM	backend/cpp/ds4/dsml_parser.cpp	333	// Definitely a literal '<' now (no chance of more bytes arriving).	COMMENT
MEDIUM	.github/workflows/test-extra.yml	603	# harness.	COMMENT
MEDIUM	.github/workflows/test-extra.yml	621	# End-to-end transcription via the e2e-backends gRPC harness against	COMMENT
MEDIUM	.github/workflows/test-extra.yml	663	# TTSStream (PCM chunks) on the e2e-backends harness.	COMMENT
MEDIUM	.github/workflows/test-extra.yml	1102	# End-to-end TTS via the e2e-backends gRPC harness. Builds the	COMMENT
MEDIUM	.github/workflows/test-extra.yml	1122	# End-to-end transcription via the e2e-backends gRPC harness. The	COMMENT
MEDIUM	.github/workflows/test-extra.yml	1161	# End-to-end audio transform via the e2e-backends gRPC harness. The	COMMENT
MEDIUM	.github/workflows/tests-pii-ner-e2e.yml	9	# This drives the container-level harness (tests/e2e-backends) via	COMMENT
MEDIUM	.github/gallery-agent/testing.go	220	fmt.Sprintf("# %s Language Model\n\nDeveloped by %s, this model represents state-of-the-art performance in natural lan	CODE

Verbosity Indicators21 hits · 39 pts

Severity	File	Line	Snippet	Context
LOW	core/gallery/upgrade.go	296	// Step 1: Download the new backend into the tmp directory	COMMENT
LOW⚡	core/gallery/upgrade.go	314	// Step 2: Validate — check that run.sh exists in the new content	COMMENT
LOW⚡	core/gallery/upgrade.go	321	// Step 3: Atomic swap — rename current to backup, then tmp to current	COMMENT
LOW	core/gallery/upgrade.go	337	// Step 4: Write updated metadata, preserving alias from old metadata	COMMENT
LOW⚡	core/gallery/upgrade.go	370	// Step 5: Re-register backends so the model loader picks up any changes	COMMENT
LOW⚡	core/gallery/upgrade.go	375	// Step 6: Remove backup	COMMENT
LOW	core/gallery/backends.go	507	// if doesn't exist, it might be an alias, so we need to check if we have a matching alias in	COMMENT
LOW	core/services/nodes/router_test.go	978	// Step 1: cached model found on old node	COMMENT
LOW⚡	core/services/nodes/router.go	663	// Step 1: Find and atomically lock a node with this model loaded	COMMENT
LOW⚡	core/services/nodes/router.go	668	// Step 2: model not loaded — it has to be cold-loaded.	COMMENT
LOW	tests/e2e/mock_backend_test.go	445	// Step 1: tokenize	COMMENT
LOW	tests/e2e/mock_backend_test.go	465	// Step 2: detokenize the returned token IDs	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	101	# Step 1: Download / resolve model	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	112	# Step 2: Convert to f16 GGUF	COMMENT
LOW	backend/python/llama-cpp-quantization/backend.py	123	# Step 3: Quantize	COMMENT
LOW	backend/python/trl/backend.py	749	# Step 1: Merge LoRA into base model	COMMENT
LOW	backend/python/trl/backend.py	798	# Step 2: Convert to GGUF using convert_hf_to_gguf.py	COMMENT
LOW⚡	pkg/functions/parse_test.go	191	// Step 1: Replace single quotes around keys and values with double quotes	COMMENT
LOW⚡	pkg/functions/parse_test.go	193	// Step 2: Replace double quotes inside values with placeholders	COMMENT
LOW⚡	pkg/functions/parse_test.go	224	// Step 1: Replace single quotes around keys and values with double quotes	COMMENT
LOW⚡	pkg/functions/parse_test.go	226	// Step 2: Replace double quotes inside values with placeholders	COMMENT

Hallucination Indicators3 hits · 30 pts

Severity	File	Line	Snippet	Context
CRITICAL	core/http/static/assets/marked.js	6	!function(e,t){"object"==typeof exports&&"undefined"!=typeof module?t(exports):"function"==typeof define&&define.amd?def	CODE
CRITICAL	core/http/static/assets/codemirror.min.js	1	!function(e,t){"object"==typeof exports&&"undefined"!=typeof module?module.exports=t():"function"==typeof define&&define	CODE
CRITICAL	backend/cpp/llama-cpp/grpc-server.cpp	1623	if (params.speculative.has_dft() && !params.speculative.draft.mparams.path.empty()) {	CODE

Magic Placeholder Names6 hits · 30 pts

Severity	File	Line	Snippet	Context
HIGH	core/http/views/agent-task-details.html	373	-H "Authorization: Bearer YOUR_API_KEY" \	CODE
HIGH	core/http/views/agent-task-details.html	394	-H "Authorization: Bearer YOUR_API_KEY" \	CODE
HIGH	core/http/views/agent-task-details.html	417	-H "Authorization: Bearer YOUR_API_KEY" \	CODE
HIGH	core/http/views/agent-task-details.html	443	-H "Authorization: Bearer YOUR_API_KEY"</code></pre>	CODE
HIGH	docs/content/features/mcp.md	444	api_key="your-api-key"	CODE
HIGH	docs/content/getting-started/troubleshooting.md	275	-H "Authorization: Bearer YOUR_API_KEY"	CODE

Docstring Block Structure3 hits · 15 pts

Severity	File	Line	Snippet	Context
HIGH	backend/python/diffusers/diffusers_dynamic_loader.py	128	Discover all subclasses of a given base class from diffusers. This function provides a generic way to discover	STRING
HIGH	backend/python/diffusers/diffusers_dynamic_loader.py	318	Resolve a pipeline class from class_name, task, or model_id. Priority: 1. If class_name is provided, look	STRING
HIGH	backend/python/diffusers/diffusers_dynamic_loader.py	428	Load a diffusers pipeline dynamically. This function resolves the appropriate pipeline class based on the prov	STRING

Dead Code5 hits · 10 pts

Severity	File	Line	Context
MEDIUM	backend/python/mlx-distributed/backend.py	604	CODE
MEDIUM	backend/python/mlx-distributed/backend.py	605	CODE
MEDIUM	backend/python/mlx-distributed/backend.py	609	CODE
MEDIUM	backend/python/mlx-distributed/backend.py	614	CODE
MEDIUM	backend/python/mlx-distributed/backend.py	616	CODE

Example Usage Blocks4 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	docker-compose.distributed.yaml	9	# Usage:	COMMENT
LOW	scripts/install-fizzbee.sh	7	# Usage:	COMMENT
LOW	scripts/model_gallery_info.py	53	# Example usage	STRING
LOW	pkg/model/watchdog.go	79	// Example usage:	COMMENT

AI Response Leakage1 hit · 5 pts

Severity	File	Line	Snippet	Context
HIGH	docs/content/reference/architecture.md	17	As much as typical open source projects starts, I, [mudler](https://github.com/mudler/), was fiddling around with [llama	CODE

Slop Phrases2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	pkg/huggingface-api/client_test.go	693	mockReadmeContent := "# Test Model\n\nThis is a test model for demonstration purposes."	CODE
MEDIUM	pkg/huggingface-api/client_test.go	712	Expect(content).To(Equal("# Test Model\n\nThis is a test model for demonstration purposes."))	CODE

Cross-Language Confusion1 hit · 2 pts

Severity	File	Line	Snippet	Context
HIGH	backend/python/common/vllm_utils_test.py	4	cd backend/python/common && python3 -m unittest vllm_utils_test	STRING

Modern Structural Boilerplate2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	backend/python/tinygrad/tool_parsers/__init__.py	11	__all__ = ["ToolCall", "ToolParser", "resolve_parser"]	CODE
LOW	backend/python/common/vllm_utils.py	14	__all__ = [	CODE

AI Structural Patterns3 hits · 2 pts

Severity	File	Line	Context
LOW	backend/python/vllm/backend.py	319	CODE
LOW	backend/python/transformers/backend.py	481	CODE
LOW	backend/python/sglang/backend.py	254	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings