LLM inference in C/C++
This report presents the forensic synthetic code analysis of ggml-org/llama.cpp, a C++ project with 126,076 GitHub stars. SynthScan v2.0 examined 843,317 lines of code across 2454 source files, recording 2361 pattern matches distributed across 24 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).
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.
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.
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.
The scanner identified 2361 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.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/quantize/quantize.cpp | 1 | #include "llama.h" | COMMENT |
| LOW | tools/ui/embed.cpp | 1 | // llama-ui-embed: generate ui.cpp / ui.h that embed UI assets as C arrays. | COMMENT |
| LOW | tools/ui/eslint.config.js | 201 | // Blank line around if blocks (if/else and else if stay one statement). | COMMENT |
| LOW | …ols/ui/tests/client/agentic-stream.perf.svelte.test.ts | 1 | // Tier 1 perf harness for the agentic thread. | COMMENT |
| LOW | tools/ui/scripts/dev.sh | 1 | #!/bin/bash | COMMENT |
| LOW | tools/ui/src/lib/vendors/decimal.js/decimal.js | 41 | // The maximum number of significant digits of the result of a calculation or base conversion. | COMMENT |
| LOW | tools/ui/src/lib/vendors/decimal.js/decimal.js | 61 | COMMENT | |
| LOW | tools/ui/src/lib/vendors/decimal.js/decimal.js | 1181 | // If the result is known to have an infinite decimal expansion, repeat this until it is clear | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Algebra.js | 1 | /* | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Algebra.js | 4061 | } | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Algebra.js | 5561 | // if (a.isSQRT()) { | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 1 | /* | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 201 | // CoreDeps - Centralized Dependency Registry ================================== | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 9141 | // Whether to wrap numbers in brackets | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 11461 | } | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 11741 | // a = _.parse(bigDec.exp(bigDec(i).toPower(2).neg().dividedBy(bigDec(n).pow(2).plus(bigDec( | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 13721 | const ret = _.callfunction(fnName, fnArgs); | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 13741 | // else if(getter instanceof Slice) { | COMMENT |
| LOW | …ols/ui/src/lib/vendors/nerdamer-prime/nerdamer.core.js | 16461 | return symB.clone(); | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Calculus.js | 1 | /* | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Extra.js | 1 | /* | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 481 | /** @type {NerdamerSymbolType} */ (_.parse(es[1])) | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 1061 | // _.parse(delta0); | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 1441 | // Nothing we can do | COMMENT |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 1641 | * | COMMENT |
| LOW | tools/ui/src/lib/constants/auto-scroll.constants.ts | 1 | export const AUTO_SCROLL_INTERVAL = 100; | COMMENT |
| LOW | tools/ui/src/lib/utils/search-results.ts | 21 | const SEPARATOR_LINE_RE = /^\s*---\s*$/; | COMMENT |
| LOW | tools/cvector-generator/pca.hpp | 1 | #include "common.h" | COMMENT |
| LOW | tools/cvector-generator/cvector-generator.cpp | 1 | #include "ggml.h" | COMMENT |
| LOW | tools/cvector-generator/cvector-generator.cpp | 21 | #include <algorithm> | COMMENT |
| LOW | tools/mtmd/mtmd.h | 1 | #ifndef MTMD_H | COMMENT |
| LOW | tools/mtmd/mtmd.h | 141 | COMMENT | |
| LOW | tools/mtmd/mtmd.h | 181 | MTMD_API void mtmd_bitmap_set_mergeable(mtmd_bitmap * bitmap, bool mergeable); | COMMENT |
| LOW | tools/mtmd/mtmd.h | 241 | MTMD_API mtmd_input_chunk * mtmd_input_chunk_get_placeholder(const mtmd_input_chunk * chunk); | COMMENT |
| LOW | tools/mtmd/mtmd.h | 281 | // the default marker is defined by mtmd_default_marker() | COMMENT |
| LOW | tools/mtmd/mtmd.h | 321 | COMMENT | |
| LOW | tools/mtmd/mtmd.h | 421 | ///////////////////////////////////////// | COMMENT |
| LOW | tools/mtmd/mtmd-helper.cpp | 1 | // fix problem with std::min and std::max | COMMENT |
| LOW | tools/mtmd/mtmd-helper.cpp | 21 | //#define MTMD_AUDIO_DEBUG | COMMENT |
| LOW | tools/mtmd/mtmd-helper.cpp | 41 | COMMENT | |
| LOW | tools/mtmd/clip-graph.h | 1 | #pragma once | COMMENT |
| LOW | tools/mtmd/tests.sh | 141 | # to test the huge models, run: ./tests.sh huge | COMMENT |
| LOW | tools/mtmd/mtmd.cpp | 1 | #include "clip.h" | COMMENT |
| LOW | tools/mtmd/mtmd.cpp | 21 | #include <cerrno> | COMMENT |
| LOW | tools/mtmd/mtmd-helper-common.h | 1 | #pragma once | COMMENT |
| LOW | tools/mtmd/clip.cpp | 1 | #include "clip.h" | COMMENT |
| LOW | tools/mtmd/clip.cpp | 4481 | nelem += img.nx() * img.ny() * 3; | COMMENT |
| LOW | tools/mtmd/mtmd-image.cpp | 261 | if (val < 0) return 0; | COMMENT |
| LOW | tools/mtmd/mtmd-helper.h | 1 | #ifndef MTMD_HELPER_H | COMMENT |
| LOW | tools/mtmd/mtmd-helper.h | 61 | mtmd_context * ctx, | COMMENT |
| LOW | tools/mtmd/mtmd-helper.h | 81 | struct mtmd_helper_init_opt opt); | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 1 | #pragma once | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 21 | #define NOMINMAX | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 41 | #define KEY_N_FF "clip.%s.feed_forward_length" | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 61 | #define KEY_PROJ_SCALE_FACTOR "clip.vision.projector.scale_factor" | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 81 | #define KEY_A_NUM_MEL_BINS "clip.audio.num_mel_bins" | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 101 | // | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 121 | #define TN_FFN_GATE "%s.blk.%d.ffn_gate.%s" | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 141 | #define TN_MM_POST_NORM "mm.post_norm.%s" | COMMENT |
| LOW | tools/mtmd/clip-impl.h | 161 | COMMENT | |
| 922 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/ui/tests/unit/agentic-strip.test.ts | 11 | function stripLegacyContextMarkers(content: string): string { | CODE |
| LOW | tools/ui/scripts/vite-plugin-splash-screen.ts | 22 | export function generateSplashScreenLinks(outDir: string): string[] { | CODE |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 353 | (function extendSimplifyForEquations() { | CODE |
| LOW | tools/ui/src/lib/vendors/nerdamer-prime/Solve.js | 355 | _.functions.simplify[0] = function simplifyWithEquationSupport(symbol) { | CODE |
| LOW | tools/ui/src/lib/contexts/chat-form-actions.context.ts | 10 | export function setChatFormActionsContext(ctx: ChatFormActionsContext): ChatFormActionsContext { | CODE |
| LOW | tools/ui/src/lib/contexts/chat-form-actions.context.ts | 17 | export function getChatFormActionsContext(): ChatFormActionsContext { | CODE |
| LOW | …ls/ui/src/lib/contexts/chat-message-actions.context.ts | 10 | export function setChatMessageActionsContext( | CODE |
| LOW | …ls/ui/src/lib/contexts/chat-message-actions.context.ts | 19 | export function getChatMessageActionsContext(): ChatMessageActionsContext { | CODE |
| LOW | tools/ui/src/lib/contexts/chat-message-edit.context.ts | 10 | export function setChatMessageEditContext(ctx: ChatMessageEditContext): ChatMessageEditContext { | CODE |
| LOW | tools/ui/src/lib/contexts/chat-message-edit.context.ts | 17 | export function getChatMessageEditContext(): ChatMessageEditContext { | CODE |
| LOW | tools/ui/src/lib/constants/browser-info.ts | 24 | export function buildBrowserInfoToolDefinition(): OpenAIToolDefinition { | CODE |
| LOW | tools/ui/src/lib/constants/read-media.ts | 39 | export function buildReadMediaToolDefinition( | CODE |
| LOW | tools/ui/src/lib/constants/get-datetime.ts | 6 | export function buildGetDatetimeToolDefinition(): OpenAIToolDefinition { | CODE |
| LOW | tools/ui/src/lib/stores/chat/context-stats.svelte.ts | 46 | function summarizeAssistantTimings(messages: DatabaseMessage[]): AssistantTimingsSummary { | CODE |
| LOW | …/ui/src/lib/stores/conversations/preferences.svelte.ts | 23 | function loadReasoningEffortDefault(): ReasoningEffort { | CODE |
| LOW | …/ui/src/lib/stores/conversations/preferences.svelte.ts | 36 | function saveReasoningEffortDefault(effort: ReasoningEffort): void { | CODE |
| LOW | …/ui/src/lib/stores/conversations/preferences.svelte.ts | 67 | function buildDisabledToolCategories(conv: DatabaseConversation | null): Set<ToolSource> { | CODE |
| LOW | tools/ui/src/lib/utils/parse-exec-shell-error.ts | 1 | export function parseExecShellCommandError( | CODE |
| LOW | tools/ui/src/lib/utils/sandbox-tool.ts | 25 | export function buildSandboxToolDefinition(includeSymbolicMath: boolean): OpenAIToolDefinition { | CODE |
| LOW | tools/ui/src/lib/utils/audio-recording.ts | 255 | export function isAudioRecordingSupported(): boolean { | CODE |
| LOW | tools/ui/src/lib/utils/jpeg-orientation.ts | 12 | export function getJpegOrientationFromDataURL(base64UrlJpeg: string): number { | CODE |
| LOW | tools/ui/src/lib/utils/sanitize.ts | 21 | export function sanitizeKeyValuePairValue(raw: string): string { | CODE |
| LOW | tools/ui/src/lib/utils/mention-token.ts | 72 | export function takeMentionDismissSnapshot( | CODE |
| LOW | tools/ui/src/lib/utils/modality-file-validation.ts | 17 | export function isFileTypeSupportedByModel( | CODE |
| LOW | tools/ui/src/lib/utils/modality-file-validation.ts | 137 | export function generateModalityErrorMessage( | CODE |
| LOW | tools/ui/src/lib/utils/clipboard.ts | 108 | export function formatMessageForClipboard( | CODE |
| LOW | tools/ui/src/lib/utils/clipboard.ts | 265 | function isValidMcpPromptAttachment(obj: unknown): obj is { | CODE |
| LOW | tools/ui/src/lib/utils/glob-search.ts | 74 | export async function runGlobSearchWithChildren( | CODE |
| LOW | tools/ui/src/lib/utils/parse-exec-shell-status.ts | 21 | export function parseExecShellCommandExitStatus( | CODE |
| LOW | tools/ui/src/lib/utils/text.ts | 15 | export function generateConversationTitle(content: string, useFirstLine: boolean = false): string { | CODE |
| LOW | tools/ui/src/lib/utils/mcp.ts | 53 | export function detectMcpTransportFromUrl(url: string): MCPTransportType { | CODE |
| LOW | tools/ui/src/lib/utils/mcp.ts | 402 | export function getMcpServerFaviconFallback(serverUrl: string): string | null { | CODE |
| LOW | …ls/ui/src/lib/utils/chat-template-thinking-detector.ts | 67 | export function detectThinkingSupportWithReason(t: string): { supported: boolean; reason: string } { | CODE |
| LOW | tools/ui/src/lib/utils/code.ts | 30 | export function splitGluedClosingCodeFences(markdown: string): string { | CODE |
| LOW | tools/ui/src/lib/utils/code.ts | 147 | export function detectIncompleteCodeBlock(markdown: string): IncompleteCodeBlock | null { | CODE |
| LOW | tools/ui/src/lib/utils/image-error-fallback.ts | 5 | export function getImageErrorFallbackHtml(src: string): string { | CODE |
| LOW | tools/ui/src/lib/utils/command-token.ts | 27 | export function takeCommandDismissSnapshot(value: string): CommandDismissSnapshot | null { | CODE |
| LOW | tools/ui/src/lib/utils/convert-files-to-extra.ts | 31 | export async function parseFilesToMessageExtras( | CODE |
| LOW | …ols/ui/src/lib/utils/chat-form-input-rich-tokenizer.ts | 705 | export function stripBlockBoundaryLineBreaks(root: HTMLElement): boolean { | CODE |
| LOW | tools/ui/src/lib/utils/file-type.ts | 112 | export function getFileTypeCategoryByExtension(filename: string): FileTypeCategory | null { | CODE |
| LOW | tools/ui/src/lib/utils/process-uploaded-files.ts | 54 | export async function processFilesToChatUploaded( | CODE |
| LOW | tools/ui/src/lib/utils/request-helpers.ts | 87 | export function formatDiagnosticErrorMessage(error: unknown): string { | CODE |
| LOW | tools/ui/src/lib/utils/attachment-display.ts | 49 | export function getAttachmentDisplayItems( | CODE |
| LOW | …ttings/SettingsChat/SettingsChatImportExportTab.svelte | 44 | function handleSettingsExportConfirm() { | CODE |
| LOW | …ttings/SettingsChat/SettingsChatImportExportTab.svelte | 69 | function handleSettingsExportCancel() { | CODE |
| LOW | …s/app/chat/ChatMessages/ChatMessage/ChatMessage.svelte | 379 | function handleShowDeleteDialogChange(show: boolean) { | CODE |
| LOW | …sage/ChatMessageToolCall/parsers/exec-shell-command.ts | 16 | export function parseExecShellCommandMeta(section: AgenticSection): ExecShellCommandMeta | null { | CODE |
| LOW | …i/src/lib/components/app/chat/ChatForm/ChatForm.svelte | 174 | async function handleWorkingDirectoryChange(newDir: string | null) { | CODE |
| LOW | …p/chat/ChatForm/ChatFormInput/ChatFormInputRich.svelte | 97 | function highlightCodeBlockElement(el: HTMLElement): boolean { | CODE |
| LOW | …p/chat/ChatForm/ChatFormInput/ChatFormInputRich.svelte | 142 | function rehighlightCaretCodeBlock() { | CODE |
| LOW | …p/chat/ChatForm/ChatFormInput/ChatFormInputRich.svelte | 467 | function moveCaretBeforeLeadingCodeBlock(key: string, extend: boolean): boolean { | CODE |
| LOW | …c/lib/components/app/chat/ChatScreen/ChatScreen.svelte | 226 | function handleErrorDialogOpenChange(open: boolean) { | CODE |
| LOW | …tion/SidebarNavigation/SidebarNavigationActions.svelte | 63 | function handleSearchModeDeactivate() { | CODE |
| LOW | …nts/app/content/MarkdownContent/MarkdownContent.svelte | 303 | function handlePreviewDialogOpenChange(open: boolean) { | CODE |
| LOW | …nts/app/content/MarkdownContent/MarkdownContent.svelte | 672 | function handleMermaidPreviewOpenChange(open: boolean) { | CODE |
| LOW | …nents/app/content/MarkdownContent/markdown-handlers.ts | 72 | export function createHandlePreviewDialogOpenChange(previewState: PreviewState) { | CODE |
| LOW | …nents/app/content/MarkdownContent/markdown-handlers.ts | 73 | return function handlePreviewDialogOpenChange(open: boolean) { | CODE |
| LOW | …nents/app/content/MarkdownContent/markdown-handlers.ts | 182 | export function createHandleMermaidPreviewOpenChange(mermaidState: MermaidPreviewState) { | CODE |
| LOW | …nents/app/content/MarkdownContent/markdown-handlers.ts | 183 | return function handleMermaidPreviewOpenChange(open: boolean) { | CODE |
| LOW | …nents/app/content/MarkdownContent/markdown-handlers.ts | 251 | export function createSetupCodeBlockActions( | CODE |
| 374 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM⚡ | tools/mtmd/clip.cpp | 4489 | // ├─────┤ │ | COMMENT |
| MEDIUM⚡ | tools/mtmd/clip.cpp | 4491 | // ├─────┤ │ | COMMENT |
| MEDIUM⚡ | tools/mtmd/clip.cpp | 4493 | // └─────┘ │ | COMMENT |
| MEDIUM⚡ | tools/mtmd/clip.cpp | 4494 | // ──────┘ x B | COMMENT |
| MEDIUM | docs/backend/OPENVINO.md | 240 | # ============================================ | COMMENT |
| MEDIUM | docs/backend/OPENVINO.md | 242 | # ============================================ | COMMENT |
| MEDIUM⚡ | docs/backend/OPENVINO.md | 271 | # ============================================ | COMMENT |
| MEDIUM⚡ | docs/backend/OPENVINO.md | 273 | # ============================================ | COMMENT |
| MEDIUM⚡ | docs/backend/OPENVINO.md | 279 | # ============================================ | COMMENT |
| MEDIUM⚡ | docs/backend/OPENVINO.md | 282 | # ============================================ | COMMENT |
| MEDIUM | docs/backend/OPENVINO.md | 309 | # ============================================ | COMMENT |
| MEDIUM | docs/backend/OPENVINO.md | 311 | # ============================================ | COMMENT |
| MEDIUM | …com/arm/aichat/internal/gguf/GgufMetadataReaderImpl.kt | 120 | // ── 1. header ────────────────────────────────────────────────────────── | COMMENT |
| MEDIUM | …com/arm/aichat/internal/gguf/GgufMetadataReaderImpl.kt | 129 | // ── 3. build structured object ──────────────────────────────────────── | COMMENT |
| MEDIUM | examples/llama.android/lib/src/main/cpp/CMakeLists.txt | 14 | # -------------------------------------------------------------------------- | COMMENT |
| MEDIUM | examples/llama.android/lib/src/main/cpp/CMakeLists.txt | 16 | # -------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 17 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 19 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 77 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 79 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 242 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 244 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 273 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 275 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 1076 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 1078 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 1089 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-function-call.py | 1091 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 29 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 31 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 97 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 99 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 254 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 256 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 337 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 339 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 372 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 374 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 913 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 915 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 927 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-parallel-tc.py | 929 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 29 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 31 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 100 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 102 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 258 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 260 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 355 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 357 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 970 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 972 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 984 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/server-test-structured.py | 986 | # --------------------------------------------------------------------------- | COMMENT |
| MEDIUM | scripts/jinja/jinja-tester.py | 157 | # ------------------------ | COMMENT |
| MEDIUM | scripts/jinja/jinja-tester.py | 159 | # ------------------------ | COMMENT |
| MEDIUM | scripts/jinja/jinja-tester.py | 287 | # ------------------------ | COMMENT |
| MEDIUM | scripts/jinja/jinja-tester.py | 289 | # ------------------------ | COMMENT |
| MEDIUM | scripts/snapdragon/qdc/run_qdc_jobs.py | 121 | # ============================================================================= | COMMENT |
| MEDIUM | scripts/snapdragon/qdc/run_qdc_jobs.py | 123 | # ============================================================================= | COMMENT |
| 19 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | convert_hf_to_gguf_update.py | 272 | except Exception as e: | STRING |
| LOW | convert_hf_to_gguf_update.py | 307 | except Exception as e: | STRING |
| LOW | tools/server/bench/bench.py | 59 | except Exception: | CODE |
| LOW | tools/server/bench/bench.py | 85 | except Exception: | CODE |
| LOW | tools/server/bench/speed-bench/speed_bench.py | 252 | except Exception as exc: | CODE |
| LOW | tools/server/bench/speed-bench/speed_bench.py | 389 | except Exception as exc: | CODE |
| LOW | tools/server/bench/speed-bench/speed_bench_compare.py | 73 | except Exception as exc: | CODE |
| LOW | tools/server/tests/utils.py | 334 | except Exception as e: | CODE |
| LOW | tools/server/tests/utils.py | 361 | except Exception as e: | CODE |
| MEDIUM | tools/server/tests/utils.py | 362 | print(f"Error waiting for server: {e}") | CODE |
| LOW | tools/server/tests/utils.py | 674 | except Exception as e: | CODE |
| MEDIUM | tools/server/tests/utils.py | 670 | def worker(index, func, args): | CODE |
| LOW | tools/server/tests/unit/test_tools_builtin.py | 163 | except Exception as e: | CODE |
| LOW | tools/server/tests/unit/test_mcp_servers.py | 674 | except Exception: | STRING |
| LOW | tools/server/tests/unit/test_router.py | 175 | except Exception as e: | CODE |
| LOW | tools/server/tests/unit/test_router.py | 509 | except Exception: | CODE |
| MEDIUM | tools/server/tests/unit/test_router.py | 172 | def _run(self, fn): | CODE |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 923 | except Exception as e: | CODE |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1588 | except Exception as e: | CODE |
| MEDIUM | examples/pydantic_models_to_grammar_examples.py | 66 | print(f"Error: unknown tool {json_data['function']}") | CODE |
| MEDIUM | examples/pydantic_models_to_grammar_examples.py | 142 | print(f"Error: unknown tool {json_data['function']}") | CODE |
| MEDIUM | examples/pydantic_models_to_grammar_examples.py | 285 | print(f"Error: unknown tool {call['function']}") | CODE |
| MEDIUM | examples/pydantic_models_to_grammar.py | 1001 | print(f"An error occurred while saving the grammar file: {e}") | CODE |
| MEDIUM | examples/pydantic_models_to_grammar.py | 1008 | print(f"An error occurred while saving the documentation file: {e}") | CODE |
| LOW | examples/llama-eval/llama-eval.py | 1055 | except Exception as e: | CODE |
| LOW | examples/llama-eval/llama-eval.py | 1097 | except Exception as e: | CODE |
| LOW | examples/llama-eval/llama-eval.py | 1137 | except Exception as e: | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1138 | print(f"Error: Cannot reach server {server_config.name} ({server_config.url}): {e}", file=sys.stderr) | CODE |
| LOW | examples/llama-eval/llama-eval.py | 1235 | except Exception as e: | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1471 | print(f"Error: --server ({len(server_urls)} URLs) and --threads ({len(thread_counts)} values) must have the same | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1478 | print(f"Error: --server-name ({len(server_names)} names) and --server ({len(server_urls)} URLs) must have th | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1489 | print("Error: GPQA dataset requires --grader-type llm") | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1499 | print(f"Error: Model mismatch. State has '{eval_state.model_name}', but --model is '{args.model}'") | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1525 | print("Error: --grader-type llm requires --grader-model or --model") | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1537 | print("Error: No existing eval state found to resume") | CODE |
| MEDIUM | examples/llama-eval/llama-eval.py | 1543 | print("Error: --grader-type llm requires --grader-model or --model") | CODE |
| LOW | examples/llama-eval/llama-server-simulator.py | 295 | except Exception as e: | CODE |
| MEDIUM | examples/llama-eval/llama-server-simulator.py | 296 | print(f"Error processing request: {e}") | CODE |
| LOW | …ples/model-conversion/scripts/causal/compare-logits.py | 18 | except Exception as e: | CODE |
| MEDIUM | …ples/model-conversion/scripts/causal/compare-logits.py | 52 | print(f"Error: PyTorch logits file not found: {pytorch_file}") | CODE |
| MEDIUM | …ples/model-conversion/scripts/causal/compare-logits.py | 57 | print(f"Error: llama.cpp logits file not found: {llamacpp_file}") | CODE |
| MEDIUM | …mples/model-conversion/scripts/causal/run-org-model.py | 116 | 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.py | 125 | print(f"Error: Prompts file '{args.prompts_file}' not found") | CODE |
| LOW⚡ | …del-conversion/scripts/embedding/run-original-model.py | 127 | except Exception as e: | CODE |
| MEDIUM⚡ | …del-conversion/scripts/embedding/run-original-model.py | 128 | print(f"Error reading prompts file: {e}") | CODE |
| MEDIUM | …del-conversion/scripts/embedding/run-original-model.py | 139 | print("Error: Model path must be specified either via --model-path argument " | CODE |
| LOW | examples/model-conversion/scripts/utils/check-nmse.py | 172 | except Exception as e: | CODE |
| LOW | …conversion/scripts/utils/hf-add-model-to-collection.py | 31 | except Exception as e: | CODE |
| MEDIUM | …conversion/scripts/utils/hf-add-model-to-collection.py | 33 | print(f"Error: {e}") | CODE |
| LOW | …conversion/scripts/utils/hf-add-model-to-collection.py | 49 | except Exception as e: | CODE |
| LOW⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 50 | except Exception as e: | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 51 | print(f"Error reading {single_file}: {e}") | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 54 | print(f"Error: No safetensors files found in {model_path}") | CODE |
| LOW⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 111 | except Exception as e: | CODE |
| LOW⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 120 | except Exception as e: | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 121 | print(f"Error reading {single_file}: {e}") | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 124 | print(f"Error: No safetensors files found in {model_path}") | CODE |
| MEDIUM | …es/model-conversion/scripts/utils/inspect-org-model.py | 182 | print(f"Error: Could not find tensor '{tensor_name}' in model index") | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 212 | print(f"Error: Tensor '{tensor_name}' not found in {tensor_file}") | CODE |
| MEDIUM⚡ | …es/model-conversion/scripts/utils/inspect-org-model.py | 216 | print(f"Error: The file '{file_path}' was not found.") | CODE |
| 57 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | convert_lora_to_gguf.py | 58 | CODE | |
| LOW | convert_lora_to_gguf.py | 222 | CODE | |
| LOW | convert_hf_to_gguf.py | 30 | CODE | |
| LOW | convert_hf_to_gguf.py | 175 | CODE | |
| LOW | convert_llama_ggml_to_gguf.py | 204 | CODE | |
| LOW | convert_llama_ggml_to_gguf.py | 287 | CODE | |
| LOW | …egacy-models/minicpmv-convert-image-encoder-to-gguf.py | 202 | CODE | |
| LOW | …egacy-models/minicpmv-convert-image-encoder-to-gguf.py | 312 | CODE | |
| LOW | tools/server/bench/bench.py | 24 | CODE | |
| LOW | tools/server/tests/utils.py | 367 | CODE | |
| LOW | tools/server/tests/utils.py | 426 | CODE | |
| LOW | tools/server/tests/unit/test_mcp_servers.py | 626 | CODE | |
| LOW | tools/server/tests/unit/test_router.py | 33 | CODE | |
| LOW | tools/server/tests/unit/test_router.py | 487 | CODE | |
| LOW | tools/server/tests/unit/test_vision_api.py | 8 | CODE | |
| LOW | tools/server/tests/unit/test_chat_completion.py | 402 | CODE | |
| LOW | tools/server/tests/unit/test_completion.py | 465 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 113 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 334 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 442 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 1253 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 1575 | CODE | |
| LOW | gguf-py/gguf/vocab.py | 161 | CODE | |
| LOW | gguf-py/gguf/vocab.py | 183 | CODE | |
| LOW | gguf-py/gguf/metadata.py | 240 | CODE | |
| LOW | gguf-py/gguf/metadata.py | 365 | CODE | |
| LOW | gguf-py/gguf/gguf_reader.py | 62 | CODE | |
| LOW | gguf-py/gguf/gguf_reader.py | 335 | CODE | |
| LOW | gguf-py/gguf/constants.py | 5595 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_convert_endian.py | 90 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_new_metadata.py | 103 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 448 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 495 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 643 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 784 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 938 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1062 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1131 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1476 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_dump.py | 32 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_dump.py | 153 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_dump.py | 228 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_dump.py | 444 | CODE | |
| LOW | gguf-py/tests/test_quants.py | 145 | CODE | |
| LOW | gguf-py/tests/test_quants.py | 88 | CODE | |
| LOW | tests/test-tokenizer-random.py | 255 | CODE | |
| LOW | tests/test-tokenizer-random.py | 267 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 30 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 53 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 337 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 397 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 568 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 343 | CODE | |
| LOW | examples/json_schema_to_grammar.py | 421 | CODE | |
| LOW | examples/pydantic_models_to_grammar_examples.py | 89 | CODE | |
| LOW | examples/convert_legacy_llama.py | 699 | CODE | |
| LOW | examples/convert_legacy_llama.py | 1100 | CODE | |
| LOW | examples/convert_legacy_llama.py | 773 | CODE | |
| LOW | examples/pydantic_models_to_grammar.py | 55 | CODE | |
| LOW | examples/pydantic_models_to_grammar.py | 280 | CODE | |
| 134 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | convert_lora_to_gguf.py | 4 | CODE | |
| LOW | convert_hf_to_gguf.py | 4 | CODE | |
| LOW | convert_llama_ggml_to_gguf.py | 2 | CODE | |
| LOW | …legacy-models/glmedge-convert-image-encoder-to-gguf.py | 8 | CODE | |
| LOW | …ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py | 8 | CODE | |
| LOW | …egacy-models/minicpmv-convert-image-encoder-to-gguf.py | 414 | CODE | |
| LOW | tools/server/bench/bench.py | 1 | CODE | |
| LOW | tools/server/bench/speed-bench/speed_bench.py | 2 | CODE | |
| LOW | tools/server/bench/speed-bench/speed_bench_compare.py | 2 | CODE | |
| LOW | tools/server/bench/speed-bench/speed_bench_compare.py | 9 | CODE | |
| LOW | tools/server/tests/conftest.py | 2 | CODE | |
| LOW | tools/server/tests/utils.py | 17 | CODE | |
| LOW | tools/server/tests/utils.py | 17 | CODE | |
| LOW | tools/server/tests/unit/test_infill.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_basic.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_security.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_tools_builtin.py | 6 | CODE | |
| LOW | tools/server/tests/unit/test_sleep.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_metrics.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_mcp_servers.py | 19 | CODE | |
| LOW | tools/server/tests/unit/test_mcp_servers.py | 21 | CODE | |
| LOW | tools/server/tests/unit/test_router.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_ctx_shift.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_speculative.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_compat_oai_responses.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_tool_call.py | 10 | CODE | |
| LOW | tools/server/tests/unit/test_vision_api.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_rerank.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_embedding.py | 5 | CODE | |
| LOW | tools/server/tests/unit/test_ignore_eos.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_chat_completion.py | 3 | CODE | |
| LOW | tools/server/tests/unit/test_tokenize.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_kv_keep_only_active.py | 4 | CODE | |
| LOW | tools/server/tests/unit/test_compat_gcp.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_proxy.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_slot_save.py | 2 | CODE | |
| LOW | tools/server/tests/unit/test_completion.py | 7 | CODE | |
| LOW | tools/server/tests/unit/test_compat_anthropic.py | 6 | CODE | |
| LOW | tools/server/tests/unit/test_template.py | 13 | CODE | |
| LOW | tools/server/tests/unit/test_stream.py | 7 | CODE | |
| LOW | tools/server/tests/unit/test_lora.py | 2 | CODE | |
| LOW | gguf-py/gguf/gguf_writer.py | 1 | CODE | |
| LOW | gguf-py/gguf/vocab.py | 1 | CODE | |
| LOW | gguf-py/gguf/metadata.py | 1 | CODE | |
| LOW | gguf-py/gguf/gguf_reader.py | 5 | CODE | |
| LOW | gguf-py/gguf/tensor_mapping.py | 1 | CODE | |
| LOW | gguf-py/gguf/utility.py | 1 | CODE | |
| LOW | gguf-py/gguf/constants.py | 1 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 1 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 2 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 3 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 4 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 5 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 6 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 7 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 8 | CODE | |
| LOW | gguf-py/gguf/__init__.py | 9 | CODE | |
| LOW | gguf-py/gguf/lazy.py | 1 | CODE | |
| LOW | gguf-py/gguf/quants.py | 1 | CODE | |
| LOW | gguf-py/gguf/scripts/gguf_convert_endian.py | 2 | CODE | |
| 115 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/ui/scripts/make-icons-circular.js | 89 | // Step 1: Scale source relative to circle diameter (not full icon), composite centered onto white canvas of full size | COMMENT |
| LOW | tools/ui/scripts/make-icons-circular.js | 101 | // Step 2: Composite scaled image onto white background, then apply circular mask | COMMENT |
| LOW | tools/ui/src/lib/utils/latex-protection.ts | 206 | // Step 0: Temporarily remove blockquote markers (>) to process LaTeX correctly | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 224 | // Step 1: Protect code blocks | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 233 | // Step 2: Protect existing LaTeX expressions | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 272 | // Step 3: Escape standalone $ before digits (currency like $5 → \$5) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 276 | // Step 4: Restore protected LaTeX expressions (they are valid) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 295 | // Step 5: Apply additional escaping functions (brackets and mhchem) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 303 | // Step 6: Convert remaining \(...\) → $...$, \[...\] → $$...$$ | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 321 | // Step 7: Restore code blocks | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 327 | // Step 8: Restore blockquote markers | COMMENT |
| LOW | tools/mtmd/models/mobilenetv5.cpp | 217 | // Step 1: Compute Q @ K.T | COMMENT |
| LOW | include/llama.h | 1215 | /// NOTE: This function does not use a jinja parser. It only support a pre-defined list of template. See more: https | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1974 | // Step 1: create input tensors that don't depend on any other tensors: | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1981 | // Step 2: use the op that you want to test in the GGML compute graph. | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1985 | // Step 3: return the output tensor. | COMMENT |
| LOW⚡ | docs/development/debugging-tests.md | 41 | #### Step 1: Reset and Setup folder context | COMMENT |
| LOW⚡ | docs/development/debugging-tests.md | 49 | #### Step 2: Setup Build Environment and Compile Test Binaries | COMMENT |
| LOW⚡ | docs/development/debugging-tests.md | 58 | #### Step 3: Find all tests available that matches REGEX | COMMENT |
| LOW | docs/development/debugging-tests.md | 86 | #### Step 4: Identify Test Command for Debugging | COMMENT |
| LOW | docs/development/debugging-tests.md | 92 | #### Step 5: Run GDB on test command | COMMENT |
| LOW | scripts/check-requirements.sh | 8 | # WARNING: This is quite IO intensive, because a fresh venv is set up for every | COMMENT |
| LOW | scripts/debug-test.sh | 62 | # Step 0: Check the args | COMMENT |
| LOW | scripts/debug-test.sh | 95 | # Step 1: Reset and Setup folder context | COMMENT |
| LOW⚡ | scripts/debug-test.sh | 109 | # Step 2: Setup Build Environment and Compile Test Binaries | COMMENT |
| LOW⚡ | scripts/debug-test.sh | 118 | # Step 3: Find all tests available that matches REGEX | COMMENT |
| LOW | scripts/debug-test.sh | 134 | # Step 4: Identify Test Command for Debugging | COMMENT |
| LOW | scripts/debug-test.sh | 171 | # Step 5: Execute or GDB Debug | COMMENT |
| LOW | skills/code-review/SKILL.md | 17 | ## Step 0 - Scope the diff and pick the checklists | COMMENT |
| LOW | skills/add-new-model/SKILL.md | 21 | ## Step 0 - Scope and dedup check | COMMENT |
| LOW⚡ | skills/add-new-model/SKILL.md | 33 | ## Step 1 - Convert the model to GGUF | COMMENT |
| LOW⚡ | skills/add-new-model/SKILL.md | 39 | ## Step 2 - Define the architecture in llama.cpp | COMMENT |
| LOW⚡ | skills/add-new-model/SKILL.md | 45 | ## Step 3 - Build the GGML graph | COMMENT |
| LOW⚡ | skills/add-new-model/SKILL.md | 51 | ## Step 4 - Optional: multimodal encoder | COMMENT |
| LOW⚡ | skills/add-new-model/SKILL.md | 57 | ## Step 5 - Optional: chat template / parsing support | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 84 | // Step 1: sum for mean | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 110 | // Step 2: compute (x - mean) → dst, accumulate variance | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 143 | // Step 3: apply scale to centered values in dst | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 100 | // Step 1: Compute sum of squares using 8-wide vectors | COMMENT |
| LOW⚡ | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 124 | // Step 2: scale = rsqrt(mean + eps) | COMMENT |
| LOW⚡ | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 132 | // Step 3: Apply scaling: broadcast scale once, reuse across loop | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c | 177 | // Step 1: sa = dot(a, state_prev[i]) | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c | 196 | // Step 2: state update + result accumulation | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 243 | // Step 1: Apply decay gate to state | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 248 | // Step 2: delta = (v - S @ k) * beta | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 255 | // Step 3: state += outer(delta, k) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 261 | // Step 4: attn_out = S @ q * scale | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 25 | // Step 1: Calculate mean along the last dimension | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 30 | // Step 2: Calculate (input - mean) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 33 | // Step 3: Calculate squared differences (input - mean)^2 | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 37 | // Step 4: Calculate variance = mean((input - mean)^2) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 41 | // Step 5: Get epsilon from op_params | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 45 | // Step 6: Calculate std = sqrt(variance + eps) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 49 | // Step 7: Normalize: output = (input - mean) / std | COMMENT |
| LOW⚡ | ggml/src/ggml-metal/CMakeLists.txt | 87 | # Step 1: concatenate shared headers + this kernel source | COMMENT |
| LOW⚡ | ggml/src/ggml-metal/CMakeLists.txt | 89 | # Step 2: remove internal #include and #pragma once | COMMENT |
| LOW⚡ | ggml/src/ggml-metal/CMakeLists.txt | 91 | # Step 3: inline ggml-common.h (replacing __embed_ggml-common.h__ sentinel) | COMMENT |
| LOW⚡ | ggml/src/ggml-metal/CMakeLists.txt | 93 | # Step 4: inline ggml-metal-impl.h | COMMENT |
| LOW⚡ | ggml/src/ggml-metal/CMakeLists.txt | 95 | # Step 5: emit an asm chunk with kind-specific start/end symbols | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 2809 | // Step 5: multiply by attn_factor | COMMENT |
| 18 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | app/download.cpp | 66 | if (!params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL | app/download.cpp | 67 | printf("%s\n", params.speculative.draft.mparams.path.c_str()); | CODE |
| CRITICAL | tests/test-model-resolution.cpp | 383 | REQUIRE(params.speculative.draft.mparams.path.empty()); | CODE |
| CRITICAL | common/arg.cpp | 404 | if (!params.speculative.draft.mparams.hf_repo.empty()) { | CODE |
| CRITICAL | common/arg.cpp | 508 | if (!params.speculative.draft.mparams.url.empty()) { | CODE |
| CRITICAL | common/arg.cpp | 536 | if (!params.speculative.draft.mparams.hf_file.empty()) { | CODE |
| CRITICAL | common/arg.cpp | 565 | if (spec_types_is_default(params) && !params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL⚡ | common/arg.cpp | 580 | if (params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL⚡ | common/arg.cpp | 590 | if (params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL⚡ | common/arg.cpp | 600 | if (params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL⚡ | common/arg.cpp | 610 | if (params.speculative.draft.mparams.path.empty()) { | CODE |
| CRITICAL | examples/speculative/speculative.cpp | 53 | if (params.speculative.draft.mparams.path.empty()) { | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …legacy-models/glmedge-convert-image-encoder-to-gguf.py | 0 | returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control charac | STRING |
| HIGH | …ls/mtmd/legacy-models/convert_image_encoder_to_gguf.py | 0 | returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control charac | STRING |
| HIGH | …egacy-models/minicpmv-convert-image-encoder-to-gguf.py | 0 | returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control charac | STRING |
| HIGH | gguf-py/gguf/vocab.py | 0 | returns list of utf-8 byte and a mapping to unicode strings. we specifically avoids mapping to whitespace/control charac | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "type": "integer", "minimum": 5, "maximum": 30 }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "type": "integer", "minimum": 5, "maximum": 30 }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "type": "integer", "minimum": 5, "maximum": 30 }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "type": "integer", "minimum": -123, "maximum": 42 }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "type": "integer", "minimum": -123, "maximum": 42 }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "type": "integer", "minimum": -123, "maximum": 42 }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "items": [ { "format": "date" }, { "format": "uuid" }, { "format": "time" }, { "format": "date-time" } ] }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "type": "string", "minlength": 1, "maxlength": 4 }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "type": "string", "minlength": 1, "maxlength": 4 }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "type": "string", "minlength": 1, "maxlength": 4 }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "enum": ["red", "amber", "green", null, 42, ["foo"]] }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "enum": ["red", "amber", "green", null, 42, ["foo"]] }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "enum": ["red", "amber", "green", null, 42, ["foo"]] }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 }) | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 }) | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "items": { "type": ["number", "integer"] }, "minitems": 3, "maxitems": 5 }) | STRING |
| HIGH | tests/test-grammar-llguidance.cpp | 0 | ({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} }, | STRING |
| HIGH | tests/test-grammar-integration.cpp | 0 | ({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} }, | STRING |
| HIGH | tests/test-json-schema-to-grammar.cpp | 0 | ({ "properties": { "b": {"type": "string"}, "a": {"type": "string"}, "d": {"type": "string"}, "c": {"type": "string"} }, | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/mtmd/tests.sh | 18 | # Check if the first argument is "big", then run test with big models | COMMENT |
| LOW | tools/mtmd/tests.sh | 42 | # Check if the second argument is "flash", then enable flash attention | COMMENT |
| LOW | tools/server/tests/utils.py | 336 | # Check if process died | COMMENT |
| LOW | tools/server/tests/utils.py | 687 | # Check if there were any exceptions | COMMENT |
| LOW | tools/server/tests/unit/test_compat_anthropic.py | 295 | # Check if model used the tool (it might not always, depending on the model) | COMMENT |
| LOW | gguf-py/gguf/metadata.py | 261 | # Check if we erroneously matched against './' or '../' etc... | COMMENT |
| LOW | gguf-py/gguf/metadata.py | 457 | # Check if Hugging Face ID is present in URL | COMMENT |
| LOW | gguf-py/gguf/metadata.py | 515 | # Check if Hugging Face ID is present in URL | COMMENT |
| LOW | gguf-py/gguf/gguf_reader.py | 80 | # Check if it's unsafe to perform slice optimization on data | COMMENT |
| LOW | gguf-py/gguf/gguf_reader.py | 246 | # Check if it's a simple scalar type. | COMMENT |
| LOW | gguf-py/gguf/utility.py | 207 | # Check if we have enough data to read the metadata | COMMENT |
| LOW⚡ | gguf-py/gguf/scripts/gguf_editor_gui.py | 1150 | # Check if this is an enum field | COMMENT |
| LOW⚡ | gguf-py/gguf/scripts/gguf_editor_gui.py | 1155 | # Check if it's a name | COMMENT |
| LOW⚡ | gguf-py/gguf/scripts/gguf_editor_gui.py | 1160 | # Check if it's a number or "NAME (value)" format | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 965 | # Check if this is an enum array | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 972 | # Check if this is an enum field | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1072 | # Check if this field has an enum type | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1317 | # Check if this is one of the linked tokenizer keys | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_editor_gui.py | 1431 | # Check if key already exists | COMMENT |
| LOW | gguf-py/gguf/scripts/gguf_dump.py | 336 | # Check if new Tensor Group | COMMENT |
| LOW | tests/test-lora-conversion-inference.sh | 139 | # Print results | COMMENT |
| LOW | examples/convert_legacy_llama.py | 1200 | # Check if it's a set of safetensors files first | COMMENT |
| LOW | examples/pydantic_models_to_grammar.py | 490 | # Check if the field is optional (not required) | COMMENT |
| LOW | examples/pydantic_models_to_grammar.py | 96 | if not parts: # Check if the list is empty | CODE |
| LOW | examples/llama-eval/llama-server-simulator.py | 127 | # Check if question text is contained in the cleaned request | COMMENT |
| LOW | …mples/model-conversion/scripts/utils/perplexity-gen.sh | 16 | # Check if data/wikitext-2-raw directory exists | COMMENT |
| LOW | …odel-conversion/scripts/utils/perplexity-run-simple.sh | 15 | # Check if data/wikitext-2-raw directory exists | COMMENT |
| LOW | examples/model-conversion/scripts/utils/common.py | 65 | # Check if there's an overlap between first and last indices or if we're at the edge case of s = 2 * max_seq | COMMENT |
| LOW | …model-conversion/scripts/utils/hf-create-collection.py | 23 | # Check if HF_TOKEN is available | COMMENT |
| LOW⚡ | …mples/model-conversion/scripts/utils/semantic_check.py | 40 | # Check if this is pooled (single embedding) or per-token embeddings | COMMENT |
| LOW | …mples/model-conversion/scripts/utils/semantic_check.py | 85 | # Check if the distance of each token embedding from the origin and compare | COMMENT |
| LOW | scripts/release.sh | 77 | # Check if we have the latest from master (skip in dry-run) | COMMENT |
| LOW | scripts/compare-llama-bench.py | 187 | # Check if all required Python libraries are installed. Would have failed earlier if not. | STRING |
| LOW⚡ | scripts/verify-checksum-models.py | 31 | # Check if the hash list file exists | COMMENT |
| LOW | scripts/verify-checksum-models.py | 54 | # Check if the file exists | COMMENT |
| LOW | scripts/verify-checksum-models.py | 70 | # Add the results to the array | COMMENT |
| LOW | scripts/verify-checksum-models.py | 82 | # Output the results as a table | COMMENT |
| LOW | scripts/apple/validate-macos.sh | 688 | # Check if authentication credentials are provided | COMMENT |
| LOW⚡ | scripts/apple/validate-macos.sh | 707 | # Check if app was created successfully | COMMENT |
| LOW⚡ | scripts/apple/validate-macos.sh | 715 | # Check if app binary exists and is executable | COMMENT |
| LOW⚡ | scripts/apple/validate-macos.sh | 723 | # Check if framework was properly embedded | COMMENT |
| LOW⚡ | scripts/apple/validate-macos.sh | 731 | # Check if framework binary exists | COMMENT |
| LOW | scripts/apple/validate-visionos.sh | 682 | # Check if authentication credentials are provided | COMMENT |
| LOW⚡ | scripts/apple/validate-visionos.sh | 707 | # Check if validation failed because the app isn't in App Store Connect | COMMENT |
| LOW⚡ | scripts/apple/validate-visionos.sh | 716 | # Check if IPA was created successfully | COMMENT |
| LOW⚡ | scripts/apple/validate-visionos.sh | 724 | # Check if app binary exists and is executable | COMMENT |
| LOW⚡ | scripts/apple/validate-visionos.sh | 732 | # Check if framework was properly embedded | COMMENT |
| LOW⚡ | scripts/apple/validate-visionos.sh | 740 | # Check if framework binary exists | COMMENT |
| LOW | scripts/apple/validate-ios.sh | 691 | # Check if authentication credentials are provided | COMMENT |
| LOW⚡ | scripts/apple/validate-ios.sh | 716 | # Check if validation failed because the app isn't in App Store Connect | COMMENT |
| LOW⚡ | scripts/apple/validate-ios.sh | 725 | # Check if IPA was created successfully | COMMENT |
| LOW⚡ | scripts/apple/validate-ios.sh | 733 | # Check if app binary exists and is executable | COMMENT |
| LOW⚡ | scripts/apple/validate-ios.sh | 741 | # Check if framework was properly embedded | COMMENT |
| LOW⚡ | scripts/apple/validate-ios.sh | 749 | # Check if framework binary exists | COMMENT |
| LOW | scripts/apple/validate-tvos.sh | 684 | # Check if authentication credentials are provided | COMMENT |
| LOW⚡ | scripts/apple/validate-tvos.sh | 709 | # Check if validation failed because the app isn't in App Store Connect | COMMENT |
| LOW⚡ | scripts/apple/validate-tvos.sh | 718 | # Check if IPA was created successfully | COMMENT |
| LOW⚡ | scripts/apple/validate-tvos.sh | 726 | # Check if app binary exists and is executable | COMMENT |
| LOW⚡ | scripts/apple/validate-tvos.sh | 734 | # Check if framework was properly embedded | COMMENT |
| LOW⚡ | scripts/apple/validate-tvos.sh | 742 | # Check if framework binary exists | COMMENT |
| 7 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/ui/scripts/make-icons-circular.js | 89 | // Step 1: Scale source relative to circle diameter (not full icon), composite centered onto white canvas of full size | COMMENT |
| LOW | tools/ui/scripts/make-icons-circular.js | 101 | // Step 2: Composite scaled image onto white background, then apply circular mask | COMMENT |
| LOW | tools/ui/src/lib/utils/latex-protection.ts | 206 | // Step 0: Temporarily remove blockquote markers (>) to process LaTeX correctly | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 224 | // Step 1: Protect code blocks | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 233 | // Step 2: Protect existing LaTeX expressions | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 272 | // Step 3: Escape standalone $ before digits (currency like $5 → \$5) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 276 | // Step 4: Restore protected LaTeX expressions (they are valid) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 295 | // Step 5: Apply additional escaping functions (brackets and mhchem) | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 303 | // Step 6: Convert remaining \(...\) → $...$, \[...\] → $$...$$ | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 321 | // Step 7: Restore code blocks | COMMENT |
| LOW⚡ | tools/ui/src/lib/utils/latex-protection.ts | 327 | // Step 8: Restore blockquote markers | COMMENT |
| LOW | tools/mtmd/models/mobilenetv5.cpp | 217 | // Step 1: Compute Q @ K.T | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1974 | // Step 1: create input tensors that don't depend on any other tensors: | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1981 | // Step 2: use the op that you want to test in the GGML compute graph. | COMMENT |
| LOW⚡ | tests/test-backend-ops.cpp | 1985 | // Step 3: return the output tensor. | COMMENT |
| LOW | scripts/debug-test.sh | 62 | # Step 0: Check the args | COMMENT |
| LOW | scripts/debug-test.sh | 95 | # Step 1: Reset and Setup folder context | COMMENT |
| LOW⚡ | scripts/debug-test.sh | 109 | # Step 2: Setup Build Environment and Compile Test Binaries | COMMENT |
| LOW⚡ | scripts/debug-test.sh | 118 | # Step 3: Find all tests available that matches REGEX | COMMENT |
| LOW | scripts/debug-test.sh | 134 | # Step 4: Identify Test Command for Debugging | COMMENT |
| LOW | scripts/debug-test.sh | 171 | # Step 5: Execute or GDB Debug | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 84 | // Step 1: sum for mean | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 110 | // Step 2: compute (x - mean) → dst, accumulate variance | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/norm_f32.c | 143 | // Step 3: apply scale to centered values in dst | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 100 | // Step 1: Compute sum of squares using 8-wide vectors | COMMENT |
| LOW⚡ | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 124 | // Step 2: scale = rsqrt(mean + eps) | COMMENT |
| LOW⚡ | ggml/src/ggml-et/et-kernels/src/rms_norm_f32.c | 132 | // Step 3: Apply scaling: broadcast scale once, reuse across loop | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c | 177 | // Step 1: sa = dot(a, state_prev[i]) | COMMENT |
| LOW | ggml/src/ggml-et/et-kernels/src/rwkv_wkv7_f32.c | 196 | // Step 2: state update + result accumulation | COMMENT |
| LOW | ggml/src/ggml-webgpu/ggml-webgpu.cpp | 1437 | // For set rows specifically, we need to check if src and idx are empty | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 243 | // Step 1: Apply decay gate to state | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 248 | // Step 2: delta = (v - S @ k) * beta | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 255 | // Step 3: state += outer(delta, k) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/gated_delta_net.cpp | 261 | // Step 4: attn_out = S @ q * scale | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 25 | // Step 1: Calculate mean along the last dimension | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 30 | // Step 2: Calculate (input - mean) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 33 | // Step 3: Calculate squared differences (input - mean)^2 | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 37 | // Step 4: Calculate variance = mean((input - mean)^2) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 41 | // Step 5: Get epsilon from op_params | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 45 | // Step 6: Calculate std = sqrt(variance + eps) | COMMENT |
| LOW⚡ | ggml/src/ggml-openvino/openvino/op/norm.cpp | 49 | // Step 7: Normalize: output = (input - mean) / std | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 2809 | // Step 5: multiply by attn_factor | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 2826 | // Step 6: repeat | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3090 | // Step 1: Prepare trans tensors for F16 type conversion to F32 if needed | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3115 | // Step 2: Prepare head tensors for tail splitting if needed | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3144 | // Step 3: Execute RotaryPositionEmbedding | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3191 | // Step 4: Copy unrotated tail portion from source to destination | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3220 | // Step 5: Cast back to F16 if needed | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3906 | // Step 1: cast the src0 (Query) to fp16 if needed | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3929 | // Step 2: create the acl tensors for src1 (Key), src2 (Value), | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 3973 | // Step 3: create the PSEShift tensor if needed | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 4042 | // Step 4: set the inputs for FusedInferAttention. | COMMENT |
| LOW | ggml/src/ggml-cann/aclnn_ops.cpp | 4103 | // Step 6: post-processing — slice padded output and/or cast to f32 | COMMENT |
| LOW | src/llama-sampler.cpp | 3402 | // Step 1: Look for restart sequences to limit the maximum repetition length. | COMMENT |
| LOW | src/llama-sampler.cpp | 3463 | // Step 2: Iterate in reverse over the last N tokens of the context, using the "Z-algorithm" (in | COMMENT |
| LOW | src/llama-sampler.cpp | 3527 | // Step 3: Iterate over dry_repeat_count and last_tokens, examining the maximum repeat length | COMMENT |
| LOW | src/llama-sampler.cpp | 3554 | // Step 4: Apply logit penalties based on the maximum repeat length for relevant tokens. | COMMENT |
| LOW | src/models/kimi-linear.cpp | 197 | // Step 1: Q, K, V projections -> [d_inner, n_tokens] | COMMENT |
| LOW⚡ | src/models/kimi-linear.cpp | 377 | // Step 1: Q projection and reshape | COMMENT |
| LOW⚡ | src/models/kimi-linear.cpp | 382 | // Step 2: KV compression | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | build-xcframework.sh | 406 | # Create a separate directory for dSYMs for all platforms | COMMENT |
| MEDIUM | build-xcframework.sh | 414 | # Create a copy of the binary that will be stripped | COMMENT |
| MEDIUM | tools/server/tests/unit/test_mcp_servers.py | 495 | # Create a minimal server that returns empty tools list | COMMENT |
| MEDIUM | ci/run.sh | 772 | # Create a fresh python3 venv and enter it | COMMENT |
| MEDIUM | gguf-py/gguf/metadata.py | 66 | # Create a new Metadata instance | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 42 | # Define the tokenizer keys that should be edited together | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 228 | # Create a dialog with enum options | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 539 | # Create a widget to hold the button | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 579 | # Create a dialog with enum options | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 705 | # Create a dialog with enum options | COMMENT |
| MEDIUM | gguf-py/gguf/scripts/gguf_editor_gui.py | 1271 | # Create a dialog with enum options | COMMENT |
| MEDIUM | examples/ts-type-to-grammar.sh | 10 | # Create a temporary directory | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 186 | # Define the rule identifier based on max_digit and min_digit | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 241 | # Define the integer part rule | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 248 | # Define the fractional part rule based on precision constraints | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 264 | # Define the float rule | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 370 | # Defining the union grammar rule separately | COMMENT |
| MEDIUM | examples/pydantic_models_to_grammar.py | 1179 | # Creating the dynamic model | COMMENT |
| MEDIUM | …model-conversion/scripts/utils/hf-create-collection.py | 39 | # Create the collection | COMMENT |
| MEDIUM⚡ | scripts/verify-checksum-models.py | 25 | # Define the path to the llama directory (parent folder of script directory) | COMMENT |
| MEDIUM⚡ | scripts/verify-checksum-models.py | 28 | # Define the file with the list of hashes and filenames | COMMENT |
| MEDIUM⚡ | scripts/verify-checksum-models.py | 40 | # Create an array to store the results | COMMENT |
| MEDIUM | scripts/apple/validate-macos.sh | 574 | # Create a simple xcscheme file to avoid xcodebuild scheme issues | COMMENT |
| MEDIUM | scripts/apple/validate-macos.sh | 670 | # Create a zip file for potential distribution | COMMENT |
| MEDIUM | scripts/apple/validate-visionos.sh | 568 | # Create a simple xcscheme file to avoid xcodebuild scheme issues | COMMENT |
| MEDIUM | scripts/apple/validate-ios.sh | 577 | # Create a simple xcscheme file to avoid xcodebuild scheme issues | COMMENT |
| MEDIUM | scripts/apple/validate-tvos.sh | 570 | # Create a simple xcscheme file to avoid xcodebuild scheme issues | COMMENT |
| MEDIUM | .github/actions/windows-setup-rocm/action.yml | 17 | # Create the venv directly at the cache location to avoid relocation issues | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/ui/tests/unit/partial-tool-call-cleanup.test.ts | 23 | // on Stop/Send immediately, mirroring what the agentic flow already does in | COMMENT |
| MEDIUM | tools/ui/tests/unit/partial-tool-call-cleanup.test.ts | 67 | // After Stop, savePartialResponseIfNeeded clears toolCalls and the agentic | COMMENT |
| MEDIUM | tools/ui/tests/unit/agentic-hotpath.bench.ts | 1 | // Microbenchmarks for the functions on the agentic-thread streaming hot path. | COMMENT |
| MEDIUM | tools/ui/tests/unit/agentic-hotpath.bench.ts | 7 | // Run: npx vitest bench --project=unit tests/unit/agentic-hotpath.bench.ts | COMMENT |
| MEDIUM | tools/ui/tests/client/README-perf.md | 5 | ## Tier 1 - `agentic-stream.perf.svelte.test.ts` (project: `client`, real Chromium) | COMMENT |
| MEDIUM | tools/ui/tests/client/README-perf.md | 52 | ## Tier 2 - `../unit/agentic-hotpath.bench.ts` (project: `unit`, node) | COMMENT |
| MEDIUM | …ols/ui/tests/client/agentic-stream.perf.svelte.test.ts | 1 | // Tier 1 perf harness for the agentic thread. | COMMENT |
| MEDIUM | …ols/ui/tests/client/agentic-stream.perf.svelte.test.ts | 11 | // Run: npx vitest --project=client --run tests/client/agentic-stream.perf.svelte.test.ts | COMMENT |
| MEDIUM | …ols/ui/tests/client/agentic-stream.perf.svelte.test.ts | 380 | // Baseline: plain text streaming, nothing agentic. | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/index.svelte.ts | 621 | // If agentic loop is running, inject as a steering message instead of starting a new flow | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/index.svelte.ts | 628 | // If non-agentic streaming is active, queue as a pending message to send after completion | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/index.svelte.ts | 1102 | // Update the first assistant message with cumulative agentic timings | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/index.svelte.ts | 1155 | // Generate LLM based title for new conversations after agentic flow completes | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/context-stats.svelte.ts | 101 | // A running agentic flow stamps llm totals on messages only when it | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/context-stats.svelte.ts | 120 | // Agentic sessions stamp the same agentic.llm totals onto every | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/agentic/index.svelte.ts | 318 | // Ensure server tools are fetched before checking if agentic is enabled | COMMENT |
| MEDIUM | tools/ui/src/lib/utils/agentic.ts | 478 | // Walk consecutive trailing tool results. The agentic loop only emits tool | COMMENT |
| MEDIUM | tools/ui/src/lib/utils/index.ts | 351 | // Audio format helper (used by agentic store and chat service) | COMMENT |
| MEDIUM | …b/components/app/chat/ChatMessages/ChatMessages.svelte | 131 | // Build display entries, grouping agentic sessions into single entries. | COMMENT |
| MEDIUM | …b/components/app/chat/ChatMessages/ChatMessages.svelte | 132 | // An agentic session = assistant(with tool_calls) → tool → assistant → tool → ... → assistant(final) | COMMENT |
| MEDIUM | …ges/ChatMessage/ChatMessageUser/ChatMessageUser.svelte | 34 | // For agentic turns, prefer the cumulative agentic.llm totals over per-call timings. | COMMENT |
| MEDIUM | …MessageAssistant/ChatMessageAssistantStatistics.svelte | 16 | // A running agentic flow stamps per-turn timings on its root message at each | COMMENT |
| MEDIUM | …MessageAssistant/ChatMessageAssistantStatistics.svelte | 17 | // turn boundary and the cumulative agentic totals only on exit; while it runs, | COMMENT |
| MEDIUM | …i/src/lib/components/app/chat/ChatForm/ChatForm.svelte | 155 | // policy-aware, same rule as the agentic flow: MCP category on and at | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | tools/ui/package-lock.json | 14790 | "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz", | CODE |
| MEDIUM | tools/ui/src/lib/vendors/nerdamer-prime/Algebra.js | 3243 | return symbol; // Absolutely nothing to do | CODE |
| MEDIUM | tools/ui/src/lib/vendors/nerdamer-prime/Algebra.js | 3754 | // this will also need utilize big number at some point | COMMENT |
| MEDIUM | tools/ui/src/lib/stores/chat/streams.svelte.ts | 343 | // the lookup by message id is robust to reordering of activeMessages, two parallel attaches | COMMENT |
| MEDIUM | gguf-py/gguf/metadata.py | 394 | # (Definitely not part of hugging face formal/informal standard) | COMMENT |
| MEDIUM | gguf-py/gguf/metadata.py | 410 | # (Definitely not part of hugging face formal/informal standard... but with model_ appended to match their s | COMMENT |
| MEDIUM | gguf-py/gguf/lazy.py | 49 | # TODO: make this even more comprehensive | COMMENT |
| LOW | examples/model-conversion/scripts/utils/common.py | 70 | # If there's overlap, just use the combined unique indices | COMMENT |
| MEDIUM | examples/embedding/embedding.cpp | 122 | // utilize the full context | COMMENT |
| LOW | scripts/jinja/jinja-tester.py | 116 | # Malformed template, just add the token | COMMENT |
| LOW | scripts/jinja/jinja-tester.py | 135 | # Malformed template, just add the token | COMMENT |
| MEDIUM | .github/workflows/release.yml | 924 | # TODO: these jobs need to use llvm toolchain in order to utilize the ccache | COMMENT |
| LOW | conversion/deepseek.py | 47 | # @bluebread: there's no window_size in config but just add it here anyway | COMMENT |
| LOW | conversion/granite.py | 766 | # If not layer id, just use the projector index | COMMENT |
| MEDIUM | ggml/src/ggml-cpu/amx/mmq.cpp | 603 | // pack again with 128 to fully utilize vector length | COMMENT |
| MEDIUM | ggml/src/ggml-cpu/amx/mmq.cpp | 645 | // pack again with 128 to fully utilize vector length | COMMENT |
| MEDIUM | ggml/src/ggml-cpu/amx/mmq.cpp | 743 | // pack again with 128 to fully utilize vector length | COMMENT |
| MEDIUM | src/llama-context.cpp | 665 | // need to implement a more robust mechanism that tries a few different inputs and analyzes the results | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/completion/README.md | 90 | <!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually --> | CODE |
| HIGH | tools/server/README.md | 28 | <!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually --> | CODE |
| HIGH | tools/cli/README.md | 7 | <!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually --> | CODE |
| HIGH | examples/gen-docs/gen-docs.cpp | 14 | #define NOTE_MESSAGE "<!-- IMPORTANT: The list below is auto-generated by llama-gen-docs; do NOT modify it manually | COMMENT |
| HIGH | ggml/src/ggml-virtgpu/regenerate_remoting.py | 3 | # Generated by Claude AI | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/server/server-schema.cpp | 71 | // TODO: implement t_max_prompt_ms | COMMENT |
| LOW | tools/server/server-task.h | 67 | int64_t t_max_prompt_ms = -1; // TODO: implement | CODE |
| LOW | tests/test-backend-ops.cpp | 8264 | // GGML_TYPE_TQ1_0, // TODO: implement for all backends | COMMENT |
| LOW | tests/test-backend-ops.cpp | 8292 | // GGML_TYPE_TQ1_0, // TODO: implement for all backends | COMMENT |
| LOW | common/speculative.cpp | 1827 | // TODO: implement | COMMENT |
| LOW | common/speculative.cpp | 1875 | // TODO: implement | COMMENT |
| LOW | common/speculative.cpp | 2033 | // TODO: implement | COMMENT |
| LOW | common/speculative.cpp | 2195 | // TODO: implement | COMMENT |
| LOW | conversion/gemma.py | 568 | # TODO: implement self.prediction_coefs.weight.clamp_(...) | COMMENT |
| LOW | ggml/src/ggml.c | 3900 | // TODO: implement non F32 return | COMMENT |
| LOW | ggml/src/ggml.c | 3924 | // TODO: implement non F32 return | COMMENT |
| LOW | ggml/src/ggml.c | 5102 | // TODO: implement antialias for modes other than bilinear | COMMENT |
| LOW | ggml/src/ggml-zdnn/ggml-zdnn.cpp | 22 | // TODO: implement support for quantized types | COMMENT |
| LOW | src/llama-model-saver.cpp | 364 | // TODO: implement split file support | COMMENT |
| LOW | src/llama-model-saver.cpp | 414 | // TODO: implement LoRA support | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | gguf-py/gguf/gguf_writer.py | 35 | logger = logging.getLogger(__name__) | CODE |
| LOW | gguf-py/gguf/vocab.py | 47 | logger = logging.getLogger(__name__) | CODE |
| LOW | gguf-py/gguf/vocab.py | 149 | def _set_special_token(self, typ: str, tid: Any) -> None: | CODE |
| LOW | gguf-py/gguf/gguf_reader.py | 39 | logger = logging.getLogger(__name__) | CODE |
| LOW | gguf-py/gguf/lazy.py | 11 | logger = logging.getLogger(__name__) | CODE |
| LOW | gguf-py/gguf/scripts/gguf_set_metadata.py | 44 | def set_metadata(reader: GGUFReader, args: argparse.Namespace) -> None: | CODE |
| LOW | scripts/tool_bench.py | 75 | logger = logging.getLogger(__name__) | CODE |
| LOW | conversion/minimax.py | 48 | def set_vocab(self) -> None: | CODE |
| LOW | conversion/command_r.py | 81 | def _set_vocab_gpt2(self) -> None: | CODE |
| LOW | conversion/__init__.py | 11 | __all__ = [ | CODE |
| LOW | conversion/laguna.py | 24 | def set_vocab(self) -> None: | CODE |
| LOW | conversion/laguna.py | 64 | def set_gguf_parameters(self) -> None: | CODE |
| LOW | conversion/bert.py | 360 | def set_vocab(self) -> None: | CODE |
| LOW⚡ | conversion/base.py | 1787 | def _set_vocab_none(self) -> None: | CODE |
| LOW⚡ | conversion/base.py | 1790 | def _set_vocab_gpt2(self) -> None: | CODE |
| LOW⚡ | conversion/base.py | 1800 | def _set_vocab_whitespace(self) -> None: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/ui/scripts/dev.sh | 8 | # Usage: | COMMENT |
| LOW | gguf-py/examples/writer.py | 13 | # Example usage: | COMMENT |
| LOW | tests/test-tokenizer-0.sh | 3 | # Usage: | COMMENT |
| LOW | tests/test-tokenizer-random.py | 4 | # Sample usage: | COMMENT |
| LOW | examples/json_schema_pydantic_example.py | 1 | # Usage: | COMMENT |
| LOW | scripts/sync-ggml-am.sh | 5 | # Usage: | COMMENT |
| LOW | scripts/hf.sh | 5 | # Usage: | COMMENT |
| LOW | scripts/release.sh | 11 | # Usage: | COMMENT |
| LOW | ggml/src/ggml-sycl/dpct/helper.hpp | 3020 | /// Example usage: | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tests/test-tokenizer-random.py | 346 | if unicodedata.category(chr(cpt)) in ("Cn", "Cs", "Co"): # undefined, surrogates, private | CODE |
| HIGH | scripts/tool_bench.py | 10 | cmake -B build && cmake --build build --config Release -j -t llama-server | STRING |
| HIGH | scripts/snapdragon/qdc/run_qdc_jobs.py | 173 | "cd $basedir; chmod +x bin/* lib/* 2>/dev/null;", | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | tools/mtmd/mtmd-helper.h | 30 | // TODO @ngxson : allow "placeholder" bitmap output for counting tokens | COMMENT |
| LOW⚡ | tools/server/tests/unit/test_ctx_shift.py | 8 | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliq | CODE |
| LOW⚡ | tools/server/tests/unit/test_ctx_shift.py | 8 | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliq | CODE |
| LOW⚡ | tools/server/tests/unit/test_ctx_shift.py | 14 | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliq | CODE |
| LOW⚡ | tools/server/tests/unit/test_ctx_shift.py | 14 | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliq | CODE |
| LOW | src/llama-graph.cpp | 1422 | LLAMA_LOG_DEBUG("%s: can_reuse = %d\n", "placeholder", cur); | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …egacy-models/minicpmv-convert-image-encoder-to-gguf.py | 82 | CODE | |
| LOW | gguf-py/gguf/utility.py | 90 | CODE | |
| LOW | gguf-py/gguf/utility.py | 185 | CODE | |
| LOW | examples/pydantic_models_to_grammar_examples.py | 308 | CODE | |
| LOW | examples/pydantic_models_to_grammar.py | 1027 | CODE | |
| LOW | scripts/tool_bench.py | 204 | CODE | |
| LOW | conversion/gpt_oss.py | 47 | CODE | |
| LOW | conversion/base.py | 124 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | tools/server/README.md | 1607 | -H "x-api-key: your-api-key" \ | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | examples/pydantic_models_to_grammar.py | 532 | Generate GBNF Grammar from Pydantic Models. This method takes a list of Pydantic models and uses them to gener | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | .github/ISSUE_TEMPLATE/030-research.yml | 9 | 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.yml | 9 | Don't forget to [check for existing refactor issue tickets](https://github.com/ggml-org/llama.cpp/issues?q=is%3A | CODE |
| LOW | conversion/base.py | 1443 | # To avoid unexpected issues - we make sure to normalize non-normalized tokens | COMMENT |
| LOW | conversion/base.py | 2212 | # To avoid unexpected issues - we make sure to normalize non-normalized tokens | COMMENT |