Repository Analysis

ggml-org/llama.cpp

LLM inference in C/C++

3.7 Likely human-written View on GitHub

Analysis Overview

This report presents the forensic synthetic code analysis of ggml-org/llama.cpp, a C++ project with 120,290 GitHub stars. SynthScan v2.0 examined 749,370 lines of code across 2197 source files, recording 2130 pattern matches distributed across 25 syntactic categories. The overall adjusted score of 3.7 places this repository in the Likely human-written band.

The scanner applied 160+ deterministic lexical heuristics, multi-line block detectors, abstract syntax tree depth profilers, and a cross-file Jaccard similarity matrix to construct a statistically normalised synthetic code estimate. All matches are individually weighted by severity coefficient and contextual multiplier before summation, and the resulting headline score is temporally discounted to account for the repository's development history relative to the commercial emergence of large language model coding tooling (November 2022 onward).

3.7
Adjusted Score
3.7
Raw Score
100%
Time Factor
2026-07-14
Last Push
120.3K
Stars
C++
Language
749.4K
Lines of Code
2.2K
Files
2.1K
Pattern Hits
2026-07-14
Scan Date
0.02
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 5HIGH 37MEDIUM 187LOW 1901

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 2130 distinct pattern matches across 25 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Over-Commented Block923 hits · 900 pts
SeverityFileLineSnippetContext
LOWtools/quantize/quantize.cpp1#include "llama.h"COMMENT
LOWtools/ui/embed.cpp1// llama-ui-embed: generate ui.cpp / ui.h that embed UI assets as C arrays.COMMENT
LOWtools/ui/scripts/dev.sh1#!/bin/bashCOMMENT
LOWtools/ui/src/lib/constants/pwa.ts301COMMENT
LOWtools/cvector-generator/pca.hpp1#include "common.h"COMMENT
LOWtools/cvector-generator/cvector-generator.cpp1#include "ggml.h"COMMENT
LOWtools/cvector-generator/cvector-generator.cpp21#include <algorithm>COMMENT
LOWtools/mtmd/mtmd.h1#ifndef MTMD_HCOMMENT
LOWtools/mtmd/mtmd.h41# define MTMD_API __declspec(dllimport)COMMENT
LOWtools/mtmd/mtmd.h141MTMD_API bool mtmd_support_audio(const mtmd_context * ctx);COMMENT
LOWtools/mtmd/mtmd.h181// - does not hold the actual dataCOMMENT
LOWtools/mtmd/mtmd.h261COMMENT
LOWtools/mtmd/mtmd-helper.cpp1// fix problem with std::min and std::maxCOMMENT
LOWtools/mtmd/mtmd-helper.cpp21#ifndef MTMD_AUDIO_DEBUGCOMMENT
LOWtools/mtmd/tests.sh141# to test the huge models, run: ./tests.sh hugeCOMMENT
LOWtools/mtmd/mtmd.cpp1#include "clip.h"COMMENT
LOWtools/mtmd/mtmd.cpp21#include <cstdio>COMMENT
LOWtools/mtmd/clip.cpp1#include "clip.h"COMMENT
LOWtools/mtmd/clip.cpp3621 if (!imgs.is_audio) {COMMENT
LOWtools/mtmd/mtmd-image.cpp341 // Adapted from https://github.com/python-pillow/Pillow/blob/main/src/libImaging/Resample.cCOMMENT
LOWtools/mtmd/mtmd-image.cpp381 // Parameters:COMMENT
LOWtools/mtmd/mtmd-helper.h1#ifndef MTMD_HELPER_HCOMMENT
LOWtools/mtmd/mtmd-helper.h41// returns nullptr on failureCOMMENT
LOWtools/mtmd/clip-impl.h1#pragma onceCOMMENT
LOWtools/mtmd/clip-impl.h21#endifCOMMENT
LOWtools/mtmd/clip-impl.h41#define KEY_PROJ_DIM "clip.%s.projection_dim"COMMENT
LOWtools/mtmd/clip-impl.h61#define KEY_PROJ_SPATIAL_OFFSETS "clip.vision.projector.spatial_offsets"COMMENT
LOWtools/mtmd/clip-impl.h81#define KEY_A_MAX_POS_EMB "clip.audio.max_pos_emb"COMMENT
LOWtools/mtmd/clip-impl.h101#define TN_ATTN_OUTPUT "%s.blk.%d.attn_out.%s"COMMENT
LOWtools/mtmd/clip-impl.h121#define TN_MM_DOWN "mm.down.%s"COMMENT
LOWtools/mtmd/clip-impl.h141// mimicpmvCOMMENT
LOWtools/mtmd/clip-impl.h161#define TN_GLM_ADAPTER_LINEAR "adapter.linear.linear.%s"COMMENT
LOWtools/mtmd/clip-impl.h181#define TN_TOK_BOI "v.boi"COMMENT
LOWtools/mtmd/clip-impl.h201#define TN_SAM_NET "v.sam.net_%d.%s"COMMENT
LOWtools/mtmd/clip-impl.h221#define TN_ATTN_REL_POS_EMB "%s.blk.%d.attn_rel_pos_emb"COMMENT
LOWtools/mtmd/clip-impl.h241#define TN_MULTI_PROJ_QUERY "%s.proj_blk.%d.query"COMMENT
LOWtools/mtmd/clip-impl.h261// mobilenetv5 (gemma3n) definitionsCOMMENT
LOWtools/mtmd/clip-impl.h281#define TN_MNV5_BLK_LAYER_SCALE "v.blk.%d.%d.layer_scale.gamma"COMMENT
LOWtools/mtmd/clip-impl.h301// gemma4COMMENT
LOWtools/mtmd/mtmd-audio.cpp1#include "mtmd-audio.h"COMMENT
LOWtools/mtmd/mtmd-cli.cpp1#include "arg.h"COMMENT
LOW…ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py281 block_count = v_hparams["num_hidden_layers"] - 1 if has_llava_projector else v_hparams["num_hidden_layers"]COMMENT
LOW…ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py301 # [COMMENT
LOW…egacy-models/minicpmv-convert-image-encoder-to-gguf.py1# coding=utf-8COMMENT
LOWtools/mtmd/models/mimovl.cpp1#include "models.h"COMMENT
LOWtools/mtmd/debug/mtmd-debug.cpp1#include "mtmd-debug.h"COMMENT
LOWtools/completion/completion.cpp1#include "arg.h"COMMENT
LOWtools/completion/completion.cpp21#include <unistd.h>COMMENT
LOWtools/server/server-common.h1#pragma onceCOMMENT
LOWtools/server/server-http.cpp1#include "common.h"COMMENT
LOWtools/server/server-task.cpp1#include "server-task.h"COMMENT
LOWtools/server/server-models.h1#pragma onceCOMMENT
LOWtools/server/server-models.h241 // note: predicate is called while holding the lockCOMMENT
LOWtools/server/server-common.cpp1#include "common.h"COMMENT
LOWtools/server/server-context.cpp1#include "server-context.h"COMMENT
LOWtools/server/server-context.cpp21#include <cstddef>COMMENT
LOWtools/server/server-http.h1#pragma onceCOMMENT
LOWtools/server/server-chat.cpp281 if (response_body.contains("max_output_tokens")) {COMMENT
LOWtools/server/server-models.cpp1#include "server-common.h"COMMENT
LOWtools/server/server-models.cpp21#include <condition_variable>COMMENT
863 more matches not shown…
Hyper-Verbose Identifiers328 hits · 315 pts
SeverityFileLineSnippetContext
LOWtools/ui/tests/unit/agentic-strip.test.ts11function stripLegacyContextMarkers(content: string): string {CODE
LOWtools/ui/scripts/vite-plugin-splash-screen.ts17export function generateSplashScreenLinks(outDir: string): string[] {CODE
LOW…ls/ui/src/lib/contexts/chat-settings-config.context.ts12export function setChatSettingsConfigContext(CODE
LOW…ls/ui/src/lib/contexts/chat-settings-config.context.ts18export function getChatSettingsConfigContext(): ChatSettingsConfigContext {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts995export function agenticPendingPermissionRequest(conversationId: string) {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1003export function agenticPendingContinueRequest(conversationId: string) {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1011export function agenticHasPendingSteeringMessage(conversationId: string) {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1015export function agenticInjectSteeringMessage(CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1023export function agenticPendingSteeringMessageContent(conversationId: string) {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1027export function agenticPendingSteeringMessageExtras(conversationId: string) {CODE
LOWtools/ui/src/lib/stores/agentic.svelte.ts1031export function agenticClearSteeringMessage(conversationId: string) {CODE
LOWtools/ui/src/lib/utils/audio-recording.ts249export function isAudioRecordingSupported(): boolean {CODE
LOWtools/ui/src/lib/utils/jpeg-orientation.ts22export function getJpegOrientationFromDataURL(base64UrlJpeg: string): number {CODE
LOWtools/ui/src/lib/utils/sanitize.ts21export function sanitizeKeyValuePairValue(raw: string): string {CODE
LOWtools/ui/src/lib/utils/modality-file-validation.ts17export function isFileTypeSupportedByModel(CODE
LOWtools/ui/src/lib/utils/modality-file-validation.ts134export function generateModalityErrorMessage(CODE
LOWtools/ui/src/lib/utils/clipboard.ts103export function formatMessageForClipboard(CODE
LOWtools/ui/src/lib/utils/clipboard.ts255function isValidMcpPromptAttachment(obj: unknown): obj is {CODE
LOWtools/ui/src/lib/utils/text.ts15export function generateConversationTitle(content: string, useFirstLine: boolean = false): string {CODE
LOWtools/ui/src/lib/utils/mcp.ts44export function detectMcpTransportFromUrl(url: string): MCPTransportType {CODE
LOW…ls/ui/src/lib/utils/chat-template-thinking-detector.ts66export function detectThinkingSupportWithReason(t: string): { supported: boolean; reason: string } {CODE
LOWtools/ui/src/lib/utils/code.ts51export function detectIncompleteCodeBlock(markdown: string): IncompleteCodeBlock | null {CODE
LOWtools/ui/src/lib/utils/image-error-fallback.ts5export function getImageErrorFallbackHtml(src: string): string {CODE
LOWtools/ui/src/lib/utils/agentic.ts181export function parseToolResultWithImages(CODE
LOWtools/ui/src/lib/utils/convert-files-to-extra.ts30export async function parseFilesToMessageExtras(CODE
LOWtools/ui/src/lib/utils/file-type.ts112export function getFileTypeCategoryByExtension(filename: string): FileTypeCategory | null {CODE
LOWtools/ui/src/lib/utils/process-uploaded-files.ts52export async function processFilesToChatUploaded(CODE
LOWtools/ui/src/lib/utils/request-helpers.ts87export function formatDiagnosticErrorMessage(error: unknown): string {CODE
LOWtools/ui/src/lib/utils/attachment-display.ts47export function getAttachmentDisplayItems(CODE
LOW…ttings/SettingsChat/SettingsChatImportExportTab.svelte44 function handleSettingsExportConfirm() {CODE
LOW…ttings/SettingsChat/SettingsChatImportExportTab.svelte68 function handleSettingsExportCancel() {CODE
LOW…s/app/chat/ChatMessages/ChatMessage/ChatMessage.svelte326 function handleShowDeleteDialogChange(show: boolean) {CODE
LOW…i/src/lib/components/app/chat/ChatForm/ChatForm.svelte412 function handleInlineResourcePickerClose() {CODE
LOW…i/src/lib/components/app/chat/ChatForm/ChatForm.svelte418 function handleInlineResourceSelect() {CODE
LOW…c/lib/components/app/chat/ChatScreen/ChatScreen.svelte173 function handleErrorDialogOpenChange(open: boolean) {CODE
LOW…tion/SidebarNavigation/SidebarNavigationActions.svelte60 function handleSearchModeDeactivate() {CODE
LOW…nts/app/content/MarkdownContent/MarkdownContent.svelte294 function handlePreviewDialogOpenChange(open: boolean) {CODE
LOW…nts/app/content/MarkdownContent/MarkdownContent.svelte634 function handleMermaidPreviewOpenChange(open: boolean) {CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts62export function createHandlePreviewDialogOpenChange(previewState: PreviewState) {CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts63 return function handlePreviewDialogOpenChange(open: boolean) {CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts162export function createHandleMermaidPreviewOpenChange(mermaidState: MermaidPreviewState) {CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts163 return function handleMermaidPreviewOpenChange(open: boolean) {CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts221export function createSetupCodeBlockActions(CODE
LOW…nents/app/content/MarkdownContent/markdown-handlers.ts251export function createSetupImageErrorHandlers(CODE
LOW…ownContent/plugins/rehype/resolve-attachment-images.ts10export function rehypeResolveAttachmentImages(options: { attachments?: DatabaseMessageExtra[] }) {CODE
LOW…omponents/app/dialogs/DialogMcpResourcesBrowser.svelte128 async function handleAttachTemplateResource() {CODE
LOW…omponents/app/dialogs/DialogMcpResourcesBrowser.svelte204 function getAllResourcesFlatInTreeOrder(): MCPResourceInfo[] {CODE
LOWtools/ui/src/lib/hooks/use-auto-scroll.svelte.ts215export function createAutoScrollController(options: AutoScrollOptions = {}): AutoScrollController {CODE
LOWtools/mtmd/legacy-models/llava_surgery_v2.py65def clean_vision_tower_from_checkpoint(checkpoint_path):CODE
LOWtools/mtmd/legacy-models/llava_surgery_v2.py101def find_relevant_checkpoints(checkpoint_paths, newline_criteria, projector):CODE
LOW…ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py233def get_non_negative_vision_feature_layers(v_hparams):CODE
LOWtools/server/bench/speed-bench/speed_bench.py139def parse_completion_response(data: dict[str, Any]) -> tuple[dict[str, Any], dict[str, Any], str | None, str]:CODE
LOWtools/server/tests/conftest.py7def stop_server_after_each_test():CODE
LOWtools/server/tests/unit/test_infill.py12def test_infill_without_input_extra():CODE
LOWtools/server/tests/unit/test_infill.py24def test_infill_with_input_extra():CODE
LOWtools/server/tests/unit/test_infill.py47def test_invalid_input_extra_req(input_extra):CODE
LOWtools/server/tests/unit/test_basic.py99def test_server_model_aliases_and_tags():CODE
LOWtools/server/tests/unit/test_security.py19def test_access_public_endpoint(endpoint: str):CODE
LOWtools/server/tests/unit/test_security.py27def test_access_static_assets_without_api_key():CODE
LOWtools/server/tests/unit/test_security.py63def test_correct_api_key_anthropic_header():CODE
268 more matches not shown…
Decorative Section Separators79 hits · 255 pts
SeverityFileLineSnippetContext
MEDIUMtools/mtmd/clip.cpp3632 // ├─────┤ │COMMENT
MEDIUMtools/mtmd/clip.cpp3634 // ├─────┤ │COMMENT
MEDIUMtools/mtmd/clip.cpp3636 // └─────┘ │COMMENT
MEDIUMtools/mtmd/clip.cpp3637 // ──────┘ x BCOMMENT
MEDIUMdocs/backend/OPENVINO.md235# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md237# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md266# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md268# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md274# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md277# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md304# ============================================COMMENT
MEDIUMdocs/backend/OPENVINO.md306# ============================================COMMENT
MEDIUM…com/arm/aichat/internal/gguf/GgufMetadataReaderImpl.kt120 // ── 1. header ──────────────────────────────────────────────────────────COMMENT
MEDIUM…com/arm/aichat/internal/gguf/GgufMetadataReaderImpl.kt129 // ── 3. build structured object ────────────────────────────────────────COMMENT
MEDIUMexamples/llama.android/lib/src/main/cpp/CMakeLists.txt14# --------------------------------------------------------------------------COMMENT
MEDIUMexamples/llama.android/lib/src/main/cpp/CMakeLists.txt16# --------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py17# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py19# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py77# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py79# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py242# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py244# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py273# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py275# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py1076# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py1078# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py1089# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-function-call.py1091# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py29# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py31# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py97# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py99# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py254# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py256# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py337# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py339# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py372# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py374# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py913# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py915# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py927# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-parallel-tc.py929# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py29# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py31# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py100# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py102# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py258# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py260# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py355# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py357# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py970# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py972# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py984# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/server-test-structured.py986# ---------------------------------------------------------------------------COMMENT
MEDIUMscripts/jinja/jinja-tester.py157# ------------------------COMMENT
MEDIUMscripts/jinja/jinja-tester.py159# ------------------------COMMENT
MEDIUMscripts/jinja/jinja-tester.py287# ------------------------COMMENT
MEDIUMscripts/jinja/jinja-tester.py289# ------------------------COMMENT
MEDIUMscripts/snapdragon/qdc/run_qdc_jobs.py121# =============================================================================COMMENT
MEDIUMscripts/snapdragon/qdc/run_qdc_jobs.py123# =============================================================================COMMENT
19 more matches not shown…
Excessive Try-Catch Wrapping114 hits · 189 pts
SeverityFileLineSnippetContext
LOWconvert_hf_to_gguf_update.py271 except Exception as e:STRING
LOWconvert_hf_to_gguf_update.py306 except Exception as e:STRING
LOWtools/server/bench/bench.py59 except Exception:CODE
LOWtools/server/bench/bench.py85 except Exception:CODE
LOWtools/server/bench/speed-bench/speed_bench.py252 except Exception as exc:CODE
LOWtools/server/bench/speed-bench/speed_bench.py389 except Exception as exc:CODE
LOWtools/server/bench/speed-bench/speed_bench_compare.py73 except Exception as exc:CODE
LOWtools/server/tests/utils.py302 except Exception as e:CODE
LOWtools/server/tests/utils.py327 except Exception as e:CODE
MEDIUMtools/server/tests/utils.py328 print(f"Error waiting for server: {e}")CODE
LOWtools/server/tests/utils.py640 except Exception as e:CODE
MEDIUMtools/server/tests/utils.py636def worker(index, func, args):CODE
LOWtools/server/tests/unit/test_router.py304 except Exception:CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py923 except Exception as e:CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1588 except Exception as e:CODE
MEDIUMexamples/pydantic_models_to_grammar_examples.py66 print(f"Error: unknown tool {json_data['function']}")CODE
MEDIUMexamples/pydantic_models_to_grammar_examples.py142 print(f"Error: unknown tool {json_data['function']}")CODE
MEDIUMexamples/pydantic_models_to_grammar_examples.py285 print(f"Error: unknown tool {call['function']}")CODE
MEDIUMexamples/pydantic_models_to_grammar.py1001 print(f"An error occurred while saving the grammar file: {e}")CODE
MEDIUMexamples/pydantic_models_to_grammar.py1008 print(f"An error occurred while saving the documentation file: {e}")CODE
LOWexamples/llama-eval/llama-eval.py1055 except Exception as e:CODE
LOWexamples/llama-eval/llama-eval.py1097 except Exception as e:CODE
LOWexamples/llama-eval/llama-eval.py1137 except Exception as e:CODE
MEDIUMexamples/llama-eval/llama-eval.py1138 print(f"Error: Cannot reach server {server_config.name} ({server_config.url}): {e}", file=sys.stderr)CODE
LOWexamples/llama-eval/llama-eval.py1235 except Exception as e:CODE
MEDIUMexamples/llama-eval/llama-eval.py1471 print(f"Error: --server ({len(server_urls)} URLs) and --threads ({len(thread_counts)} values) must have the sameCODE
MEDIUMexamples/llama-eval/llama-eval.py1478 print(f"Error: --server-name ({len(server_names)} names) and --server ({len(server_urls)} URLs) must have thCODE
MEDIUMexamples/llama-eval/llama-eval.py1489 print("Error: GPQA dataset requires --grader-type llm")CODE
MEDIUMexamples/llama-eval/llama-eval.py1499 print(f"Error: Model mismatch. State has '{eval_state.model_name}', but --model is '{args.model}'")CODE
MEDIUMexamples/llama-eval/llama-eval.py1525 print("Error: --grader-type llm requires --grader-model or --model")CODE
MEDIUMexamples/llama-eval/llama-eval.py1537 print("Error: No existing eval state found to resume")CODE
MEDIUMexamples/llama-eval/llama-eval.py1543 print("Error: --grader-type llm requires --grader-model or --model")CODE
LOWexamples/llama-eval/llama-server-simulator.py295 except Exception as e:CODE
MEDIUMexamples/llama-eval/llama-server-simulator.py296 print(f"Error processing request: {e}")CODE
LOW…ples/model-conversion/scripts/causal/compare-logits.py18 except Exception as e:CODE
MEDIUM…ples/model-conversion/scripts/causal/compare-logits.py52 print(f"Error: PyTorch logits file not found: {pytorch_file}")CODE
MEDIUM…ples/model-conversion/scripts/causal/compare-logits.py57 print(f"Error: llama.cpp logits file not found: {llamacpp_file}")CODE
MEDIUM…mples/model-conversion/scripts/causal/run-org-model.py116 print("Error: Model path must be specified either via --model-path argument or MODEL_PATH environment variable")CODE
MEDIUM…del-conversion/scripts/embedding/run-original-model.py125 print(f"Error: Prompts file '{args.prompts_file}' not found")CODE
LOW…del-conversion/scripts/embedding/run-original-model.py127 except Exception as e:CODE
MEDIUM…del-conversion/scripts/embedding/run-original-model.py128 print(f"Error reading prompts file: {e}")CODE
MEDIUM…del-conversion/scripts/embedding/run-original-model.py139 print("Error: Model path must be specified either via --model-path argument "CODE
LOWexamples/model-conversion/scripts/utils/check-nmse.py172 except Exception as e:CODE
LOW…conversion/scripts/utils/hf-add-model-to-collection.py31 except Exception as e:CODE
MEDIUM…conversion/scripts/utils/hf-add-model-to-collection.py33 print(f"Error: {e}")CODE
LOW…conversion/scripts/utils/hf-add-model-to-collection.py49 except Exception as e:CODE
LOW…es/model-conversion/scripts/utils/inspect-org-model.py50 except Exception as e:CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py51 print(f"Error reading {single_file}: {e}")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py54 print(f"Error: No safetensors files found in {model_path}")CODE
LOW…es/model-conversion/scripts/utils/inspect-org-model.py111 except Exception as e:CODE
LOW…es/model-conversion/scripts/utils/inspect-org-model.py120 except Exception as e:CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py121 print(f"Error reading {single_file}: {e}")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py124 print(f"Error: No safetensors files found in {model_path}")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py182 print(f"Error: Could not find tensor '{tensor_name}' in model index")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py212 print(f"Error: Tensor '{tensor_name}' not found in {tensor_file}")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py216 print(f"Error: The file '{file_path}' was not found.")CODE
LOW…es/model-conversion/scripts/utils/inspect-org-model.py218 except Exception as e:CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py219 print(f"An error occurred: {e}")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py268 print("Error: --model-path not provided and MODEL_PATH environment variable not set")CODE
MEDIUM…es/model-conversion/scripts/utils/inspect-org-model.py273 print(f"Error: Model path does not exist: {model_path}")CODE
54 more matches not shown…
Deep Nesting182 hits · 172 pts
SeverityFileLineSnippetContext
LOWconvert_lora_to_gguf.py58CODE
LOWconvert_lora_to_gguf.py222CODE
LOWconvert_hf_to_gguf.py30CODE
LOWconvert_hf_to_gguf.py171CODE
LOWconvert_llama_ggml_to_gguf.py204CODE
LOWconvert_llama_ggml_to_gguf.py287CODE
LOW…egacy-models/minicpmv-convert-image-encoder-to-gguf.py202CODE
LOW…egacy-models/minicpmv-convert-image-encoder-to-gguf.py312CODE
LOWtools/server/bench/bench.py24CODE
LOWtools/server/tests/utils.py333CODE
LOWtools/server/tests/utils.py392CODE
LOWtools/server/tests/unit/test_router.py33CODE
LOWtools/server/tests/unit/test_router.py282CODE
LOWtools/server/tests/unit/test_vision_api.py8CODE
LOWtools/server/tests/unit/test_chat_completion.py402CODE
LOWtools/server/tests/unit/test_completion.py463CODE
LOWgguf-py/gguf/gguf_writer.py113CODE
LOWgguf-py/gguf/gguf_writer.py330CODE
LOWgguf-py/gguf/gguf_writer.py438CODE
LOWgguf-py/gguf/gguf_writer.py1128CODE
LOWgguf-py/gguf/gguf_writer.py1384CODE
LOWgguf-py/gguf/vocab.py161CODE
LOWgguf-py/gguf/vocab.py183CODE
LOWgguf-py/gguf/metadata.py240CODE
LOWgguf-py/gguf/metadata.py365CODE
LOWgguf-py/gguf/gguf_reader.py57CODE
LOWgguf-py/gguf/gguf_reader.py318CODE
LOWgguf-py/gguf/constants.py4651CODE
LOWgguf-py/gguf/scripts/gguf_convert_endian.py72CODE
LOWgguf-py/gguf/scripts/gguf_new_metadata.py103CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py448CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py495CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py643CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py784CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py938CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1062CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1131CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1476CODE
LOWgguf-py/gguf/scripts/gguf_dump.py32CODE
LOWgguf-py/gguf/scripts/gguf_dump.py153CODE
LOWgguf-py/gguf/scripts/gguf_dump.py228CODE
LOWgguf-py/gguf/scripts/gguf_dump.py444CODE
LOWgguf-py/tests/test_quants.py145CODE
LOWgguf-py/tests/test_quants.py88CODE
LOWtests/test-tokenizer-random.py255CODE
LOWtests/test-tokenizer-random.py267CODE
LOWexamples/json_schema_to_grammar.py30CODE
LOWexamples/json_schema_to_grammar.py53CODE
LOWexamples/json_schema_to_grammar.py337CODE
LOWexamples/json_schema_to_grammar.py397CODE
LOWexamples/json_schema_to_grammar.py568CODE
LOWexamples/json_schema_to_grammar.py343CODE
LOWexamples/json_schema_to_grammar.py421CODE
LOWexamples/pydantic_models_to_grammar_examples.py89CODE
LOWexamples/convert_legacy_llama.py699CODE
LOWexamples/convert_legacy_llama.py1100CODE
LOWexamples/convert_legacy_llama.py773CODE
LOWexamples/pydantic_models_to_grammar.py55CODE
LOWexamples/pydantic_models_to_grammar.py280CODE
LOWexamples/pydantic_models_to_grammar.py528CODE
122 more matches not shown…
Unused Imports161 hits · 160 pts
SeverityFileLineSnippetContext
LOWconvert_lora_to_gguf.py4CODE
LOWconvert_hf_to_gguf.py4CODE
LOWconvert_llama_ggml_to_gguf.py2CODE
LOW…legacy-models/glmedge-convert-image-encoder-to-gguf.py8CODE
LOW…ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py8CODE
LOW…egacy-models/minicpmv-convert-image-encoder-to-gguf.py414CODE
LOWtools/server/bench/bench.py1CODE
LOWtools/server/bench/speed-bench/speed_bench.py2CODE
LOWtools/server/bench/speed-bench/speed_bench_compare.py2CODE
LOWtools/server/bench/speed-bench/speed_bench_compare.py9CODE
LOWtools/server/tests/conftest.py2CODE
LOWtools/server/tests/utils.py17CODE
LOWtools/server/tests/utils.py17CODE
LOWtools/server/tests/unit/test_infill.py2CODE
LOWtools/server/tests/unit/test_basic.py3CODE
LOWtools/server/tests/unit/test_security.py3CODE
LOWtools/server/tests/unit/test_tools_builtin.py4CODE
LOWtools/server/tests/unit/test_sleep.py3CODE
LOWtools/server/tests/unit/test_router.py3CODE
LOWtools/server/tests/unit/test_ctx_shift.py2CODE
LOWtools/server/tests/unit/test_speculative.py2CODE
LOWtools/server/tests/unit/test_compat_oai_responses.py3CODE
LOWtools/server/tests/unit/test_tool_call.py10CODE
LOWtools/server/tests/unit/test_vision_api.py2CODE
LOWtools/server/tests/unit/test_rerank.py2CODE
LOWtools/server/tests/unit/test_embedding.py5CODE
LOWtools/server/tests/unit/test_ignore_eos.py2CODE
LOWtools/server/tests/unit/test_chat_completion.py3CODE
LOWtools/server/tests/unit/test_tokenize.py2CODE
LOWtools/server/tests/unit/test_kv_keep_only_active.py4CODE
LOWtools/server/tests/unit/test_compat_gcp.py2CODE
LOWtools/server/tests/unit/test_proxy.py2CODE
LOWtools/server/tests/unit/test_slot_save.py2CODE
LOWtools/server/tests/unit/test_completion.py7CODE
LOWtools/server/tests/unit/test_compat_anthropic.py6CODE
LOWtools/server/tests/unit/test_template.py13CODE
LOWtools/server/tests/unit/test_lora.py2CODE
LOWgguf-py/gguf/gguf_writer.py1CODE
LOWgguf-py/gguf/vocab.py1CODE
LOWgguf-py/gguf/metadata.py1CODE
LOWgguf-py/gguf/gguf_reader.py5CODE
LOWgguf-py/gguf/tensor_mapping.py1CODE
LOWgguf-py/gguf/utility.py1CODE
LOWgguf-py/gguf/constants.py1CODE
LOWgguf-py/gguf/__init__.py1CODE
LOWgguf-py/gguf/__init__.py2CODE
LOWgguf-py/gguf/__init__.py3CODE
LOWgguf-py/gguf/__init__.py4CODE
LOWgguf-py/gguf/__init__.py5CODE
LOWgguf-py/gguf/__init__.py6CODE
LOWgguf-py/gguf/__init__.py7CODE
LOWgguf-py/gguf/__init__.py8CODE
LOWgguf-py/gguf/__init__.py9CODE
LOWgguf-py/gguf/lazy.py1CODE
LOWgguf-py/gguf/quants.py1CODE
LOWgguf-py/gguf/scripts/gguf_convert_endian.py2CODE
LOWgguf-py/gguf/scripts/gguf_new_metadata.py2CODE
LOWgguf-py/gguf/scripts/gguf_editor_gui.py2CODE
LOWgguf-py/gguf/scripts/gguf_dump.py2CODE
LOWgguf-py/gguf/scripts/gguf_hash.py2CODE
101 more matches not shown…
Cross-File Repetition25 hits · 125 pts
SeverityFileLineSnippetContext
HIGH…legacy-models/glmedge-convert-image-encoder-to-gguf.py0returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control characSTRING
HIGH…ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py0returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control characSTRING
HIGH…egacy-models/minicpmv-convert-image-encoder-to-gguf.py0returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control characSTRING
HIGHgguf-py/gguf/vocab.py0returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control characSTRING
HIGHtests/test-grammar-llguidance.cpp0({ "type": "integer", "minimum": 5, "maximum": 30 })STRING
HIGHtests/test-grammar-integration.cpp0({ "type": "integer", "minimum": 5, "maximum": 30 })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "type": "integer", "minimum": 5, "maximum": 30 })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "type": "integer", "minimum": -123, "maximum": 42 })STRING
HIGHtests/test-grammar-integration.cpp0({ "type": "integer", "minimum": -123, "maximum": 42 })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "type": "integer", "minimum": -123, "maximum": 42 })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] })STRING
HIGHtests/test-grammar-integration.cpp0({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "type": "string", "minlength": 1, "maxlength": 4 })STRING
HIGHtests/test-grammar-integration.cpp0({ "type": "string", "minlength": 1, "maxlength": 4 })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "type": "string", "minlength": 1, "maxlength": 4 })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "enum": ["red", "amber", "green", null, 42, ["foo"]] })STRING
HIGHtests/test-grammar-integration.cpp0({ "enum": ["red", "amber", "green", null, 42, ["foo"]] })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "enum": ["red", "amber", "green", null, 42, ["foo"]] })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 })STRING
HIGHtests/test-grammar-integration.cpp0({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 })STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 })STRING
HIGHtests/test-grammar-llguidance.cpp0({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} },STRING
HIGHtests/test-grammar-integration.cpp0({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} },STRING
HIGHtests/test-json-schema-to-grammar.cpp0({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} },STRING
Structural Annotation Overuse66 hits · 121 pts
SeverityFileLineSnippetContext
LOWtools/ui/scripts/make-icons-circular.js86 // Step 1: Scale source relative to circle diameter (not full icon), composite centered onto white canvas of full sizeCOMMENT
LOWtools/ui/scripts/make-icons-circular.js100 // Step 2: Composite scaled image onto white background, then apply circular maskCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts151 // Step 0: Temporarily remove blockquote markers (>) to process LaTeX correctlyCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts165 // Step 1: Protect code blocksCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts174 // Step 2: Protect existing LaTeX expressionsCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts210 // Step 3: Escape standalone $ before digits (currency like $5 → \$5)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts214 // Step 4: Restore protected LaTeX expressions (they are valid)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts229 // Step 5: Apply additional escaping functions (brackets and mhchem)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts237 // Step 6: Convert remaining \(...\) → $...$, \[...\] → $$...$$COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts253 // Step 7: Restore code blocksCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts259 // Step 8: Restore blockquote markersCOMMENT
LOWtools/mtmd/models/mobilenetv5.cpp217 // Step 1: Compute Q @ K.TCOMMENT
LOWinclude/llama.h1185 /// NOTE: This function does not use a jinja parser. It only support a pre-defined list of template. See more: httpsCOMMENT
LOWtests/test-backend-ops.cpp1970 // Step 1: create input tensors that don't depend on any other tensors:COMMENT
LOWtests/test-backend-ops.cpp1977 // Step 2: use the op that you want to test in the GGML compute graph.COMMENT
LOWtests/test-backend-ops.cpp1981 // Step 3: return the output tensor.COMMENT
LOWdocs/development/debugging-tests.md41#### Step 1: Reset and Setup folder contextCOMMENT
LOWdocs/development/debugging-tests.md49#### Step 2: Setup Build Environment and Compile Test BinariesCOMMENT
LOWdocs/development/debugging-tests.md58#### Step 3: Find all tests available that matches REGEXCOMMENT
LOWdocs/development/debugging-tests.md86#### Step 4: Identify Test Command for DebuggingCOMMENT
LOWdocs/development/debugging-tests.md92#### Step 5: Run GDB on test commandCOMMENT
LOWscripts/check-requirements.sh8# WARNING: This is quite IO intensive, because a fresh venv is set up for everyCOMMENT
LOWscripts/debug-test.sh62# Step 0: Check the argsCOMMENT
LOWscripts/debug-test.sh95# Step 1: Reset and Setup folder contextCOMMENT
LOWscripts/debug-test.sh109# Step 2: Setup Build Environment and Compile Test BinariesCOMMENT
LOWscripts/debug-test.sh118# Step 3: Find all tests available that matches REGEXCOMMENT
LOWscripts/debug-test.sh134# Step 4: Identify Test Command for DebuggingCOMMENT
LOWscripts/debug-test.sh171# Step 5: Execute or GDB DebugCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c84 // Step 1: sum for meanCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c110 // Step 2: compute (x - mean) → dst, accumulate varianceCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c143 // Step 3: apply scale to centered values in dstCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c100 // Step 1: Compute sum of squares using 8-wide vectorsCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c124 // Step 2: scale = rsqrt(mean + eps)COMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c132 // Step 3: Apply scaling: broadcast scale once, reuse across loopCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c177 // Step 1: sa = dot(a, state_prev[i])COMMENT
LOWggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c196 // Step 2: state update + result accumulationCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp198 // Step 1: Apply decay gate to stateCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp203 // Step 2: delta = (v - S @ k) * betaCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp210 // Step 3: state += outer(delta, k)COMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp216 // Step 4: attn_out = S @ q * scaleCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp25 // Step 1: Calculate mean along the last dimensionCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp30 // Step 2: Calculate (input - mean)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp33 // Step 3: Calculate squared differences (input - mean)^2COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp37 // Step 4: Calculate variance = mean((input - mean)^2)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp41 // Step 5: Get epsilon from op_paramsCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp45 // Step 6: Calculate std = sqrt(variance + eps)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp49 // Step 7: Normalize: output = (input - mean) / stdCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp2809 // Step 5: multiply by attn_factorCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp2826 // Step 6: repeatCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3090 // Step 1: Prepare trans tensors for F16 type conversion to F32 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3115 // Step 2: Prepare head tensors for tail splitting if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3144 // Step 3: Execute RotaryPositionEmbeddingCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3191 // Step 4: Copy unrotated tail portion from source to destinationCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3220 // Step 5: Cast back to F16 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3906 // Step 1: cast the src0 (Query) to fp16 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3929 // Step 2: create the acl tensors for src1 (Key), src2 (Value),COMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3973 // Step 3: create the PSEShift tensor if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp4042 // Step 4: set the inputs for FusedInferAttention.COMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp4103 // Step 6: post-processing — slice padded output and/or cast to f32COMMENT
LOWsrc/llama-sampler.cpp2945 // Step 1: Look for restart sequences to limit the maximum repetition length.COMMENT
6 more matches not shown…
Redundant / Tautological Comments65 hits · 115 pts
SeverityFileLineSnippetContext
LOWtools/mtmd/tests.sh18# Check if the first argument is "big", then run test with big modelsCOMMENT
LOWtools/mtmd/tests.sh42# Check if the second argument is "flash", then enable flash attentionCOMMENT
LOWtools/server/tests/utils.py304 # Check if process diedCOMMENT
LOWtools/server/tests/utils.py653 # Check if there were any exceptionsCOMMENT
LOWtools/server/tests/unit/test_compat_anthropic.py295 # Check if model used the tool (it might not always, depending on the model)COMMENT
LOWtools/tts/convert_pt_to_hf.py40# Check if the loaded model is a state_dict or a model instanceCOMMENT
LOWgguf-py/gguf/metadata.py261 # Check if we erroneously matched against './' or '../' etc...COMMENT
LOWgguf-py/gguf/metadata.py457 # Check if Hugging Face ID is present in URLCOMMENT
LOWgguf-py/gguf/metadata.py515 # Check if Hugging Face ID is present in URLCOMMENT
LOWgguf-py/gguf/gguf_reader.py75 # Check if it's unsafe to perform slice optimization on dataCOMMENT
LOWgguf-py/gguf/gguf_reader.py233 # Check if it's a simple scalar type.COMMENT
LOWgguf-py/gguf/utility.py207 # Check if we have enough data to read the metadataCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1150 # Check if this is an enum fieldCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1155 # Check if it's a nameCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1160 # Check if it's a number or "NAME (value)" formatCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py965 # Check if this is an enum arrayCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py972 # Check if this is an enum fieldCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1072 # Check if this field has an enum typeCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1317 # Check if this is one of the linked tokenizer keysCOMMENT
LOWgguf-py/gguf/scripts/gguf_editor_gui.py1431 # Check if key already existsCOMMENT
LOWgguf-py/gguf/scripts/gguf_dump.py336 # Check if new Tensor GroupCOMMENT
LOWtests/test-lora-conversion-inference.sh139# Print resultsCOMMENT
LOWexamples/convert_legacy_llama.py1200 # Check if it's a set of safetensors files firstCOMMENT
LOWexamples/pydantic_models_to_grammar.py490 # Check if the field is optional (not required)COMMENT
LOWexamples/pydantic_models_to_grammar.py96 if not parts: # Check if the list is emptyCODE
LOWexamples/llama-eval/llama-server-simulator.py127 # Check if question text is contained in the cleaned requestCOMMENT
LOW…mples/model-conversion/scripts/utils/perplexity-gen.sh16# Check if data/wikitext-2-raw directory existsCOMMENT
LOW…odel-conversion/scripts/utils/perplexity-run-simple.sh15# Check if data/wikitext-2-raw directory existsCOMMENT
LOWexamples/model-conversion/scripts/utils/common.py65 # Check if there's an overlap between first and last indices or if we're at the edge case of s = 2 * max_seqCOMMENT
LOW…model-conversion/scripts/utils/hf-create-collection.py23 # Check if HF_TOKEN is availableCOMMENT
LOW…mples/model-conversion/scripts/utils/semantic_check.py40 # Check if this is pooled (single embedding) or per-token embeddingsCOMMENT
LOW…mples/model-conversion/scripts/utils/semantic_check.py85 # Check if the distance of each token embedding from the origin and compareCOMMENT
LOWscripts/compare-llama-bench.py187 # Check if all required Python libraries are installed. Would have failed earlier if not.STRING
LOWscripts/verify-checksum-models.py31# Check if the hash list file existsCOMMENT
LOWscripts/verify-checksum-models.py54 # Check if the file existsCOMMENT
LOWscripts/verify-checksum-models.py70 # Add the results to the arrayCOMMENT
LOWscripts/verify-checksum-models.py82# Output the results as a tableCOMMENT
LOWscripts/apple/validate-macos.sh688# Check if authentication credentials are providedCOMMENT
LOWscripts/apple/validate-macos.sh707# Check if app was created successfullyCOMMENT
LOWscripts/apple/validate-macos.sh715# Check if app binary exists and is executableCOMMENT
LOWscripts/apple/validate-macos.sh723# Check if framework was properly embeddedCOMMENT
LOWscripts/apple/validate-macos.sh731# Check if framework binary existsCOMMENT
LOWscripts/apple/validate-visionos.sh682# Check if authentication credentials are providedCOMMENT
LOWscripts/apple/validate-visionos.sh707# Check if validation failed because the app isn't in App Store ConnectCOMMENT
LOWscripts/apple/validate-visionos.sh716 # Check if IPA was created successfullyCOMMENT
LOWscripts/apple/validate-visionos.sh724 # Check if app binary exists and is executableCOMMENT
LOWscripts/apple/validate-visionos.sh732 # Check if framework was properly embeddedCOMMENT
LOWscripts/apple/validate-visionos.sh740 # Check if framework binary existsCOMMENT
LOWscripts/apple/validate-ios.sh691# Check if authentication credentials are providedCOMMENT
LOWscripts/apple/validate-ios.sh716# Check if validation failed because the app isn't in App Store ConnectCOMMENT
LOWscripts/apple/validate-ios.sh725 # Check if IPA was created successfullyCOMMENT
LOWscripts/apple/validate-ios.sh733 # Check if app binary exists and is executableCOMMENT
LOWscripts/apple/validate-ios.sh741 # Check if framework was properly embeddedCOMMENT
LOWscripts/apple/validate-ios.sh749 # Check if framework binary existsCOMMENT
LOWscripts/apple/validate-tvos.sh684# Check if authentication credentials are providedCOMMENT
LOWscripts/apple/validate-tvos.sh709# Check if validation failed because the app isn't in App Store ConnectCOMMENT
LOWscripts/apple/validate-tvos.sh718 # Check if IPA was created successfullyCOMMENT
LOWscripts/apple/validate-tvos.sh726 # Check if app binary exists and is executableCOMMENT
LOWscripts/apple/validate-tvos.sh734 # Check if framework was properly embeddedCOMMENT
LOWscripts/apple/validate-tvos.sh742 # Check if framework binary existsCOMMENT
5 more matches not shown…
Verbosity Indicators60 hits · 110 pts
SeverityFileLineSnippetContext
LOWtools/ui/scripts/make-icons-circular.js86 // Step 1: Scale source relative to circle diameter (not full icon), composite centered onto white canvas of full sizeCOMMENT
LOWtools/ui/scripts/make-icons-circular.js100 // Step 2: Composite scaled image onto white background, then apply circular maskCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts151 // Step 0: Temporarily remove blockquote markers (>) to process LaTeX correctlyCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts165 // Step 1: Protect code blocksCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts174 // Step 2: Protect existing LaTeX expressionsCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts210 // Step 3: Escape standalone $ before digits (currency like $5 → \$5)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts214 // Step 4: Restore protected LaTeX expressions (they are valid)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts229 // Step 5: Apply additional escaping functions (brackets and mhchem)COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts237 // Step 6: Convert remaining \(...\) → $...$, \[...\] → $$...$$COMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts253 // Step 7: Restore code blocksCOMMENT
LOWtools/ui/src/lib/utils/latex-protection.ts259 // Step 8: Restore blockquote markersCOMMENT
LOWtools/mtmd/models/mobilenetv5.cpp217 // Step 1: Compute Q @ K.TCOMMENT
LOWtests/test-backend-ops.cpp1970 // Step 1: create input tensors that don't depend on any other tensors:COMMENT
LOWtests/test-backend-ops.cpp1977 // Step 2: use the op that you want to test in the GGML compute graph.COMMENT
LOWtests/test-backend-ops.cpp1981 // Step 3: return the output tensor.COMMENT
LOWscripts/debug-test.sh62# Step 0: Check the argsCOMMENT
LOWscripts/debug-test.sh95# Step 1: Reset and Setup folder contextCOMMENT
LOWscripts/debug-test.sh109# Step 2: Setup Build Environment and Compile Test BinariesCOMMENT
LOWscripts/debug-test.sh118# Step 3: Find all tests available that matches REGEXCOMMENT
LOWscripts/debug-test.sh134# Step 4: Identify Test Command for DebuggingCOMMENT
LOWscripts/debug-test.sh171# Step 5: Execute or GDB DebugCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c84 // Step 1: sum for meanCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c110 // Step 2: compute (x - mean) → dst, accumulate varianceCOMMENT
LOWggml/src/ggml-et/et-kernels/src/norm_f32.c143 // Step 3: apply scale to centered values in dstCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c100 // Step 1: Compute sum of squares using 8-wide vectorsCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c124 // Step 2: scale = rsqrt(mean + eps)COMMENT
LOWggml/src/ggml-et/et-kernels/src/rms_norm_f32.c132 // Step 3: Apply scaling: broadcast scale once, reuse across loopCOMMENT
LOWggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c177 // Step 1: sa = dot(a, state_prev[i])COMMENT
LOWggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c196 // Step 2: state update + result accumulationCOMMENT
LOWggml/src/ggml-webgpu/ggml-webgpu.cpp1291 // For set rows specifically, we need to check if src and idx are emptyCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp198 // Step 1: Apply decay gate to stateCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp203 // Step 2: delta = (v - S @ k) * betaCOMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp210 // Step 3: state += outer(delta, k)COMMENT
LOWggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp216 // Step 4: attn_out = S @ q * scaleCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp25 // Step 1: Calculate mean along the last dimensionCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp30 // Step 2: Calculate (input - mean)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp33 // Step 3: Calculate squared differences (input - mean)^2COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp37 // Step 4: Calculate variance = mean((input - mean)^2)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp41 // Step 5: Get epsilon from op_paramsCOMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp45 // Step 6: Calculate std = sqrt(variance + eps)COMMENT
LOWggml/src/ggml-openvino/openvino/op/norm.cpp49 // Step 7: Normalize: output = (input - mean) / stdCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp2809 // Step 5: multiply by attn_factorCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp2826 // Step 6: repeatCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3090 // Step 1: Prepare trans tensors for F16 type conversion to F32 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3115 // Step 2: Prepare head tensors for tail splitting if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3144 // Step 3: Execute RotaryPositionEmbeddingCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3191 // Step 4: Copy unrotated tail portion from source to destinationCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3220 // Step 5: Cast back to F16 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3906 // Step 1: cast the src0 (Query) to fp16 if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3929 // Step 2: create the acl tensors for src1 (Key), src2 (Value),COMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp3973 // Step 3: create the PSEShift tensor if neededCOMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp4042 // Step 4: set the inputs for FusedInferAttention.COMMENT
LOWggml/src/ggml-cann/aclnn_ops.cpp4103 // Step 6: post-processing — slice padded output and/or cast to f32COMMENT
LOWsrc/llama-sampler.cpp2945 // Step 1: Look for restart sequences to limit the maximum repetition length.COMMENT
LOWsrc/llama-sampler.cpp3006 // Step 2: Iterate in reverse over the last N tokens of the context, using the "Z-algorithm" (inCOMMENT
LOWsrc/llama-sampler.cpp3070 // Step 3: Iterate over dry_repeat_count and last_tokens, examining the maximum repeat lengthCOMMENT
LOWsrc/llama-sampler.cpp3097 // Step 4: Apply logit penalties based on the maximum repeat length for relevant tokens.COMMENT
LOWsrc/models/kimi-linear.cpp197 // Step 1: Q, K, V projections -> [d_inner, n_tokens]COMMENT
LOWsrc/models/kimi-linear.cpp377 // Step 1: Q projection and reshapeCOMMENT
LOWsrc/models/kimi-linear.cpp382 // Step 2: KV compressionCOMMENT
Self-Referential Comments27 hits · 86 pts
SeverityFileLineSnippetContext
MEDIUMbuild-xcframework.sh367 # Create a separate directory for dSYMs for all platformsCOMMENT
MEDIUMbuild-xcframework.sh375 # Create a copy of the binary that will be strippedCOMMENT
MEDIUMtools/tts/convert_pt_to_hf.py135# Create the weight mapCOMMENT
MEDIUMci/run.sh714 # Create a fresh python3 venv and enter itCOMMENT
MEDIUMgguf-py/gguf/metadata.py66 # Create a new Metadata instanceCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py42# Define the tokenizer keys that should be edited togetherCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py228 # Create a dialog with enum optionsCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py539 # Create a widget to hold the buttonCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py579 # Create a dialog with enum optionsCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py705 # Create a dialog with enum optionsCOMMENT
MEDIUMgguf-py/gguf/scripts/gguf_editor_gui.py1271 # Create a dialog with enum optionsCOMMENT
MEDIUMexamples/ts-type-to-grammar.sh10# Create a temporary directoryCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py186 # Define the rule identifier based on max_digit and min_digitCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py241 # Define the integer part ruleCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py248 # Define the fractional part rule based on precision constraintsCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py264 # Define the float ruleCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py370 # Defining the union grammar rule separatelyCOMMENT
MEDIUMexamples/pydantic_models_to_grammar.py1179 # Creating the dynamic modelCOMMENT
MEDIUM…model-conversion/scripts/utils/hf-create-collection.py39 # Create the collectionCOMMENT
MEDIUMscripts/verify-checksum-models.py25# Define the path to the llama directory (parent folder of script directory)COMMENT
MEDIUMscripts/verify-checksum-models.py28# Define the file with the list of hashes and filenamesCOMMENT
MEDIUMscripts/verify-checksum-models.py40# Create an array to store the resultsCOMMENT
MEDIUMscripts/apple/validate-macos.sh574# Create a simple xcscheme file to avoid xcodebuild scheme issuesCOMMENT
MEDIUMscripts/apple/validate-macos.sh670# Create a zip file for potential distributionCOMMENT
MEDIUMscripts/apple/validate-visionos.sh568# Create a simple xcscheme file to avoid xcodebuild scheme issuesCOMMENT
MEDIUMscripts/apple/validate-ios.sh577# Create a simple xcscheme file to avoid xcodebuild scheme issuesCOMMENT
MEDIUMscripts/apple/validate-tvos.sh570# Create a simple xcscheme file to avoid xcodebuild scheme issuesCOMMENT
Hallucination Indicators5 hits · 50 pts
SeverityFileLineSnippetContext
CRITICALapp/download.cpp66 if (!params.speculative.draft.mparams.path.empty()) {CODE
CRITICALapp/download.cpp67 printf("%s\n", params.speculative.draft.mparams.path.c_str());CODE
CRITICALcommon/arg.cpp383 if (!params.speculative.draft.mparams.hf_repo.empty()) {CODE
CRITICALcommon/arg.cpp493 if (!params.speculative.draft.mparams.url.empty()) {CODE
CRITICALexamples/speculative/speculative.cpp52 if (params.speculative.draft.mparams.path.empty()) {CODE
AI Slop Vocabulary16 hits · 40 pts
SeverityFileLineSnippetContext
MEDIUMtools/ui/package-lock.json14321 "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz",CODE
MEDIUMtools/ui/src/lib/stores/chat.svelte.ts345 // the lookup by message id is robust to reordering of activeMessages, two parallel attachesCOMMENT
MEDIUMgguf-py/gguf/metadata.py394 # (Definitely not part of hugging face formal/informal standard)COMMENT
MEDIUMgguf-py/gguf/metadata.py410 # (Definitely not part of hugging face formal/informal standard... but with model_ appended to match their sCOMMENT
MEDIUMgguf-py/gguf/lazy.py49 # TODO: make this even more comprehensiveCOMMENT
LOWexamples/model-conversion/scripts/utils/common.py70 # If there's overlap, just use the combined unique indicesCOMMENT
MEDIUMexamples/embedding/embedding.cpp122 // utilize the full contextCOMMENT
LOWscripts/jinja/jinja-tester.py116 # Malformed template, just add the tokenCOMMENT
LOWscripts/jinja/jinja-tester.py135 # Malformed template, just add the tokenCOMMENT
MEDIUM.github/workflows/release.yml802 # TODO: these jobs need to use llvm toolchain in order to utilize the ccacheCOMMENT
LOWconversion/deepseek.py43 # @bluebread: there's no window_size in config but just add it here anywayCOMMENT
LOWconversion/granite.py497 # If not layer id, just use the projector indexCOMMENT
MEDIUMggml/src/ggml-zendnn/ggml-zendnn.cpp600 // gather + batched matmul + scatter approach. Future versions will leverageCOMMENT
MEDIUMggml/src/ggml-cpu/amx/mmq.cpp603 // pack again with 128 to fully utilize vector lengthCOMMENT
MEDIUMggml/src/ggml-cpu/amx/mmq.cpp645 // pack again with 128 to fully utilize vector lengthCOMMENT
MEDIUMggml/src/ggml-cpu/amx/mmq.cpp743 // pack again with 128 to fully utilize vector lengthCOMMENT
Synthetic Comment Markers7 hits · 38 pts
SeverityFileLineSnippetContext
HIGHCONTRIBUTING.md18Code that is initially generated by AI and subsequently edited will still be considered AI-generated. AI assistance is pCODE
HIGHtools/completion/README.md90<!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually -->CODE
HIGHtools/server/README.md28<!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually -->CODE
HIGHtools/cli/README.md7<!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually -->CODE
HIGHexamples/gen-docs/gen-docs.cpp14#define NOTE_MESSAGE "<!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manuallyCOMMENT
HIGHggml/src/ggml-virtgpu/regenerate_remoting.py3# Generated by Claude AISTRING
HIGHggml/src/ggml-hexagon/ggml-hexagon.cpp2064 // Check for F16 support only as requestedCOMMENT
Modern AI Meta-Vocabulary12 hits · 36 pts
SeverityFileLineSnippetContext
MEDIUMtools/ui/src/lib/stores/agentic.svelte.ts396 // Ensure built-in tools are fetched before checking if agentic is enabledCOMMENT
MEDIUMtools/ui/src/lib/stores/chat.svelte.ts968 // If agentic loop is running, inject as a steering message instead of starting a new flowCOMMENT
MEDIUMtools/ui/src/lib/stores/chat.svelte.ts974 // If non-agentic streaming is active, queue as a pending message to send after completionCOMMENT
MEDIUMtools/ui/src/lib/stores/chat.svelte.ts1174 // Update the first assistant message with cumulative agentic timingsCOMMENT
MEDIUMtools/ui/src/lib/stores/chat.svelte.ts1352 // Generate LLM based title for new conversations after agentic flow completesCOMMENT
MEDIUMtools/ui/src/lib/utils/legacy-migration.ts181 // Has agentic markers - full migration neededCOMMENT
MEDIUMtools/ui/src/lib/utils/agentic.ts271 // Walk consecutive trailing tool results. The agentic loop only emits toolCOMMENT
MEDIUM…b/components/app/chat/ChatMessages/ChatMessages.svelte183 // Build display entries, grouping agentic sessions into single entries.COMMENT
MEDIUM…b/components/app/chat/ChatMessages/ChatMessages.svelte184 // An agentic session = assistant(with tool_calls) → tool → assistant → tool → ... → assistant(final)COMMENT
MEDIUM…app/chat/ChatMessages/ChatMessageAgenticContent.svelte119 // Group flat sections into agentic turnsCOMMENT
MEDIUM…ges/ChatMessage/ChatMessageUser/ChatMessageUser.svelte60 // For agentic turns, prefer the cumulative agentic.llm totals over per-call timings.COMMENT
MEDIUMtools/ui/src/lib/hooks/use-context-gauge.svelte.ts175 // Agentic sessions stamp the same agentic.llm totals onto everyCOMMENT
TODO Padding15 hits · 22 pts
SeverityFileLineSnippetContext
LOWtools/server/server-schema.cpp71 // TODO: implement t_max_prompt_msCOMMENT
LOWtools/server/server-task.h67 int64_t t_max_prompt_ms = -1; // TODO: implementCODE
LOWtests/test-backend-ops.cpp7795 // GGML_TYPE_TQ1_0, GGML_TYPE_TQ2_0, // TODO: implement for all backendsCOMMENT
LOWtests/test-backend-ops.cpp7820 // GGML_TYPE_TQ1_0, GGML_TYPE_TQ2_0, // TODO: implement for all backendsCOMMENT
LOWcommon/speculative.cpp1671 // TODO: implementCOMMENT
LOWcommon/speculative.cpp1723 // TODO: implementCOMMENT
LOWcommon/speculative.cpp1885 // TODO: implementCOMMENT
LOWcommon/speculative.cpp2051 // TODO: implementCOMMENT
LOWconversion/gemma.py554 # TODO: implement self.prediction_coefs.weight.clamp_(...)COMMENT
LOWggml/src/ggml.c3882 // TODO: implement non F32 returnCOMMENT
LOWggml/src/ggml.c3906 // TODO: implement non F32 returnCOMMENT
LOWggml/src/ggml.c5052 // TODO: implement antialias for modes other than bilinearCOMMENT
LOWggml/src/ggml-zdnn/ggml-zdnn.cpp22 // TODO: implement support for quantized typesCOMMENT
LOWsrc/llama-model-saver.cpp304 // TODO: implement split file supportCOMMENT
LOWsrc/llama-model-saver.cpp352 // TODO: implement LoRA supportCOMMENT
Modern Structural Boilerplate13 hits · 14 pts
SeverityFileLineSnippetContext
LOWgguf-py/gguf/gguf_writer.py35logger = logging.getLogger(__name__)CODE
LOWgguf-py/gguf/vocab.py47logger = logging.getLogger(__name__)CODE
LOWgguf-py/gguf/vocab.py149 def _set_special_token(self, typ: str, tid: Any) -> None:CODE
LOWgguf-py/gguf/gguf_reader.py34logger = logging.getLogger(__name__)CODE
LOWgguf-py/gguf/lazy.py11logger = logging.getLogger(__name__)CODE
LOWgguf-py/gguf/scripts/gguf_set_metadata.py44def set_metadata(reader: GGUFReader, args: argparse.Namespace) -> None:CODE
LOWscripts/tool_bench.py75logger = logging.getLogger(__name__)CODE
LOWconversion/command_r.py76 def _set_vocab_gpt2(self) -> None:CODE
LOWconversion/__init__.py11__all__ = [CODE
LOWconversion/bert.py356 def set_vocab(self) -> None:CODE
LOWconversion/base.py1705 def _set_vocab_none(self) -> None:CODE
LOWconversion/base.py1708 def _set_vocab_gpt2(self) -> None:CODE
LOWconversion/base.py1718 def _set_vocab_whitespace(self) -> None:CODE
Cross-Language Confusion3 hits · 12 pts
SeverityFileLineSnippetContext
HIGHtests/test-tokenizer-random.py346 if unicodedata.category(chr(cpt)) in ("Cn", "Cs", "Co"): # undefined, surrogates, privateCODE
HIGHscripts/tool_bench.py10 cmake -B build && cmake --build build --config Release -j -t llama-serverSTRING
HIGHscripts/snapdragon/qdc/run_qdc_jobs.py173 "cd $basedir; chmod +x bin/* lib/* 2>/dev/null;",CODE
Example Usage Blocks8 hits · 12 pts
SeverityFileLineSnippetContext
LOWtools/ui/scripts/dev.sh8# Usage:COMMENT
LOWgguf-py/examples/writer.py13# Example usage:COMMENT
LOWtests/test-tokenizer-0.sh3# Usage:COMMENT
LOWtests/test-tokenizer-random.py4# Sample usage:COMMENT
LOWexamples/json_schema_pydantic_example.py1# Usage:COMMENT
LOWscripts/sync-ggml-am.sh5# Usage:COMMENT
LOWscripts/hf.sh5# Usage:COMMENT
LOWggml/src/ggml-sycl/dpct/helper.hpp3020 /// Example usage:COMMENT
Fake / Example Data6 hits · 8 pts
SeverityFileLineSnippetContext
LOWtools/mtmd/mtmd-helper.h130 // TODO @ngxson : allow "placeholder" bitmap output for counting tokensCOMMENT
LOWtools/server/tests/unit/test_ctx_shift.py8Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqCODE
LOWtools/server/tests/unit/test_ctx_shift.py8Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqCODE
LOWtools/server/tests/unit/test_ctx_shift.py14Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqCODE
LOWtools/server/tests/unit/test_ctx_shift.py14Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqCODE
LOWsrc/llama-graph.cpp1293 LLAMA_LOG_DEBUG("%s: can_reuse = %d\n", "placeholder", cur);CODE
AI Structural Patterns8 hits · 8 pts
SeverityFileLineSnippetContext
LOW…egacy-models/minicpmv-convert-image-encoder-to-gguf.py82CODE
LOWgguf-py/gguf/utility.py90CODE
LOWgguf-py/gguf/utility.py185CODE
LOWexamples/pydantic_models_to_grammar_examples.py308CODE
LOWexamples/pydantic_models_to_grammar.py1027CODE
LOWscripts/tool_bench.py204CODE
LOWconversion/gpt_oss.py46CODE
LOWconversion/base.py116CODE
Magic Placeholder Names1 hit · 5 pts
SeverityFileLineSnippetContext
HIGHtools/server/README.md1526 -H "x-api-key: your-api-key" \CODE
Docstring Block Structure1 hit · 5 pts
SeverityFileLineSnippetContext
HIGHexamples/pydantic_models_to_grammar.py532 Generate GBNF Grammar from Pydantic Models. This method takes a list of Pydantic models and uses them to generSTRING
Slop Phrases4 hits · 5 pts
SeverityFileLineSnippetContext
LOW.github/ISSUE_TEMPLATE/030-research.yml9 Don't forget to check for any [duplicate research issue tickets](https://github.com/ggml-org/llama.cpp/issues?q=CODE
LOW.github/ISSUE_TEMPLATE/040-refactor.yml9 Don't forget to [check for existing refactor issue tickets](https://github.com/ggml-org/llama.cpp/issues?q=is%3ACODE
LOWconversion/base.py1364 # To avoid unexpected issues - we make sure to normalize non-normalized tokensCOMMENT
LOWconversion/base.py2130 # To avoid unexpected issues - we make sure to normalize non-normalized tokensCOMMENT
Overly Generic Function Names1 hit · 1 pts
SeverityFileLineSnippetContext
LOWtools/server/tests/conftest.py19def do_something():CODE