Home / abetlen / llama-cpp-python
AI Slop Code RIsk Monitor

abetlen/llama-cpp-python

Python bindings for llama.cpp

9.9 Moderate pattern signal View on GitHub

Maintenance and Governance Review

This report summarizes scanner evidence for abetlen/llama-cpp-python, a Python project with 10,642 GitHub stars. SynthScan examined 45,738 lines of code across 96 source files, recording 538 pattern matches distributed across 16 syntactic categories. The legacy adjusted pattern signal is 9.9 (Moderate pattern signal); it is not an authorship estimate.

Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.

9.9
Adjusted Signal
9.9
Raw Pattern Score
100%
Time Factor
2026-10-01
Last Push
10.6K
Stars
Python
Language
45.7K
Lines of Code
96
Files
538
Pattern Hits
2026-10-12
Scan Date
0.17
HC Hit Rate

Risk Review · v3

Use each vector to guide manual verification of the underlying findings.

Cognitive Ownership Risk 31.0/100 · MEDIUM

Can maintainers explain intent, assumptions, and trade-offs? 455 finding(s).

Hallucinated Dependency & Supply Chain Risk 5.0/100 · LOW

Are imports declared and package/API names verified? 41 finding(s).

Agent Over-Engineering Ratio 26.2/100 · MEDIUM

Are wrappers, guards, and repeated structures necessary? 373 finding(s).

PR Review Velocity Impact 50.6/100 · HIGH

Which changes need focused human review? 538 finding(s).

What These Metrics Mean

Adjusted Signal
Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score
Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor
A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits
Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate
High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files
Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 0HIGH 16MEDIUM 14LOW 508

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 538 distinct pattern matches across 16 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. 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 configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Hyper-Verbose Identifiers315 hits · 202 pts
SeverityFileLineSnippetContext
LOW⚡tests/test_llama.py69def llama_cpp_transformer_model_path():CODE
LOW⚡tests/test_llama.py77def llama_cpp_embedding_model_path():CODE
LOW⚡tests/test_llama.py85def llama_cpp_recurrent_model_path():CODE
LOW⚡tests/test_llama.py93def llama_cpp_hybrid_model_path():CODE
LOWtests/test_llama.py23def test_llama_cpp_tokenization():CODE
LOWtests/test_llama.py270def test_real_llama_repeated_prompt_cache(llama_cpp_model_path):STRING
LOWtests/test_llama.py301def _assert_prompt_cache_reset_handles_history_edit(STRING
LOWtests/test_llama.py344def test_recurrent_model_prompt_cache_reset(llama_cpp_recurrent_model_path):STRING
LOWtests/test_llama.py352def test_hybrid_model_prompt_cache_reset(llama_cpp_hybrid_model_path):STRING
LOWtests/test_llama.py394def _eval_alternate_same_length_prompt(model, tokens, expected_next_token):STRING
LOWtests/test_llama.py418def _assert_exact_cached_prompt_reuse_matches_fresh(STRING
LOWtests/test_llama.py458def _assert_loaded_exact_cached_prompt_reuse_matches_fresh(STRING
LOWtests/test_llama.py495def _assert_ram_cache_exact_prompt_hit_matches_fresh(STRING
LOWtests/test_llama.py550def _assert_shorter_prompt_prefix_reuse_matches_fresh(STRING
LOWtests/test_llama.py578def test_exact_cached_prompt_reuse_matches_fresh(STRING
LOWtests/test_llama.py594def test_loaded_exact_cached_prompt_reuse_matches_fresh(STRING
LOWtests/test_llama.py610def test_ram_cache_exact_prompt_hit_matches_fresh(STRING
LOWtests/test_llama.py626def test_shorter_prompt_prefix_reuse_matches_fresh(STRING
LOWtests/test_llama.py639def test_real_llama_embeddings(llama_cpp_embedding_model_path):STRING
LOWtests/test_llama_grammar.py18def test_composed_pydantic_grammar():STRING
LOWtests/test_llama_speculative.py6def test_find_candidate_pred_tokens():CODE
LOWtests/test_llama_chat_format.py85def test_hf_tokenizer_config_str_to_chat_formatter():CODE
LOW⚡examples/server/server.py13183 def request_needs_prompt_logits(request: CompletionRequest) -> bool:CODE
LOW⚡examples/server/server.py13189 def request_needs_uncached_prompt_logprobs(request: CompletionRequest) -> bool:CODE
LOW⚡examples/server/server.py13193 def clear_sequence_cache_match(request: CompletionRequest) -> None:CODE
LOW⚡examples/server/server.py13197 def can_lookup_sequence_cache(self, request: CompletionRequest) -> bool:CODE
LOW⚡examples/server/server.py13204 def is_sequence_cache_match_usable(CODE
LOWexamples/server/server.py4529 def _regex_literal_prefix_and_remainder(pattern: str) -> Tuple[str, str]:CODE
LOWexamples/server/server.py4584 def _consume_optional_literal_prefix(CODE
LOWexamples/server/server.py4599 def _split_regex_alternatives(pattern: str) -> List[str]:CODE
LOWexamples/server/server.py4633 def _regex_lookahead_literal_specs(cls, pattern: str) -> List[Tuple[str, bool]]:CODE
LOWexamples/server/server.py4671 def _regex_capture_end_literal_specs(cls, pattern: str) -> List[Tuple[str, bool]]:CODE
LOWexamples/server/server.py4683 def _regex_capture_end_literals(cls, pattern: str) -> List[str]:CODE
LOWexamples/server/server.py4728 def _literal_suffix_prefix_length(text: str, literal: str) -> int:CODE
LOWexamples/server/server.py4763 def _compile_iterator_pattern(pattern: str) -> Optional[Tuple[str, str]]:CODE
LOWexamples/server/server.py4774 def _compile_iterator_block_pattern(CODE
LOWexamples/server/server.py4853 def _compile_word_capture_pattern(CODE
LOWexamples/server/server.py4872 def _consume_until_any_literal(CODE
LOWexamples/server/server.py4907 def _compile_tool_call_item_plan(CODE
LOWexamples/server/server.py4976 def _compile_segment_message_plan(CODE
LOWexamples/server/server.py5045 def _compile_tagged_message_plan(CODE
LOWexamples/server/server.py5224 def _parameter_schema_for_tool(CODE
LOWexamples/server/server.py5250 def _raw_string_tool_arguments(self, tool_name: str, value: str) -> Optional[Dict[str, str]]:CODE
LOWexamples/server/server.py5282 def _single_string_tool_argument_name(self, tool_name: str) -> Optional[str]:CODE
LOWexamples/server/server.py5314 def _text_tool_argument_from_object(CODE
LOWexamples/server/server.py5364 def _raw_object_tool_arguments(cls, value: str) -> Optional[Dict[str, Any]]:CODE
LOWexamples/server/server.py5386 def _trim_partial_gemma_quote_marker(value: str) -> str:CODE
LOWexamples/server/server.py5585 def _advance_direct_tool_call_state(self, text: str) -> Tuple[bool, List[Dict[str, Any]]]:CODE
LOWexamples/server/server.py5836 def _advance_direct_stream_state(self, text: str) -> Tuple[bool, List[Dict[str, Any]]]:CODE
LOWexamples/server/server.py6673 def _partial_regex_key_value_item(CODE
LOWexamples/server/server.py6708 def _trim_partial_tool_call_prefix(CODE
LOWexamples/server/server.py7172 def _normalize_tool_call_item(CODE
LOWexamples/server/server.py7255 def _serialize_partial_json_prefix(cls, value: Any) -> str:CODE
LOWexamples/server/server.py7302 def _contains_partial_json_value(cls, value: Any) -> bool:CODE
LOWexamples/server/server.py7314 def _serialize_partial_json_state(cls, value: Any) -> Tuple[str, bool]:CODE
LOWexamples/server/server.py7357 def _serialize_tool_arguments(cls, arguments: Any, *, partial: bool = False) -> str:CODE
LOWexamples/server/server.py7897 def _tools_for_response_parser(CODE
LOWexamples/server/server.py7933 def _response_reasoning_effort(CODE
LOWexamples/server/server.py7945 def _response_text_from_content(content: Any) -> str:CODE
LOWexamples/server/server.py7969 def _response_chat_content_from_content(CODE
255 more matches not shown…
Empty Error Handlers9 hits · 45 pts
SeverityFileLineSnippetContext
HIGHexamples/server/server.py12546except FileNotFoundError:CODE
HIGHexamples/server/server.py12554except OSError:CODE
HIGHexamples/server/server.py15022except ValueError:CODE
HIGHexamples/server/server.py16339except WebSocketDisconnect:CODE
HIGHexamples/server/server.py15631except RuntimeError:CODE
HIGHexamples/server/server.py10834except OSError:CODE
HIGHexamples/server/server.py12491except FileNotFoundError:CODE
HIGHexamples/server/server.py11229except OSError:CODE
HIGHllama_cpp/llama.py1552except UnicodeError:CODE
Deep Nesting65 hits · 40 pts
SeverityFileLineSnippetContext
LOWdocker/open_llama/hug_model.py29CODE
LOWexamples/low_level_api/low_level_api_chat_cpp.py344CODE
LOWexamples/low_level_api/low_level_api_chat_cpp.py705CODE
LOWexamples/server/server.py15764CODE
LOWexamples/server/server.py329CODE
LOWexamples/server/server.py364CODE
LOWexamples/server/server.py526CODE
LOWexamples/server/server.py1818CODE
LOWexamples/server/server.py2239CODE
LOWexamples/server/server.py3503CODE
LOWexamples/server/server.py4529CODE
LOWexamples/server/server.py5419CODE
LOWexamples/server/server.py5585CODE
LOWexamples/server/server.py5836CODE
LOWexamples/server/server.py6003CODE
LOWexamples/server/server.py6380CODE
LOWexamples/server/server.py6734CODE
LOWexamples/server/server.py7366CODE
LOWexamples/server/server.py7687CODE
LOWexamples/server/server.py8137CODE
LOWexamples/server/server.py8698CODE
LOWexamples/server/server.py8952CODE
LOWexamples/server/server.py9308CODE
LOWexamples/server/server.py9492CODE
LOWexamples/server/server.py10021CODE
LOWexamples/server/server.py11000CODE
LOWexamples/server/server.py11239CODE
LOWexamples/server/server.py11601CODE
LOWexamples/server/server.py12771CODE
LOWexamples/server/server.py13390CODE
LOWexamples/server/server.py13564CODE
LOWexamples/server/server.py13791CODE
LOWexamples/server/server.py13887CODE
LOWexamples/server/server.py14340CODE
LOWexamples/server/server.py15893CODE
LOWexamples/server/server.py15938CODE
LOWexamples/server/server.py16201CODE
LOWexamples/server/server.py330CODE
LOWexamples/server/server.py377CODE
LOWllama_cpp/llama_chat_format.py1460CODE
LOWllama_cpp/llama_chat_format.py1823CODE
LOWllama_cpp/llama_chat_format.py4132CODE
LOWllama_cpp/llama_chat_format.py1490CODE
LOWllama_cpp/llama_chat_format.py1574CODE
LOWllama_cpp/llama_chat_format.py1874CODE
LOWllama_cpp/llama_chat_format.py2139CODE
LOWllama_cpp/llama_chat_format.py2850CODE
LOWllama_cpp/llama_chat_format.py3140CODE
LOWllama_cpp/llama_chat_format.py3401CODE
LOWllama_cpp/llama_chat_format.py3674CODE
LOWllama_cpp/llama_chat_format.py1624CODE
LOWllama_cpp/llama.py60CODE
LOWllama_cpp/llama.py704CODE
LOWllama_cpp/llama.py854CODE
LOWllama_cpp/llama.py1195CODE
LOWllama_cpp/_ctypes_extensions.py26CODE
LOWllama_cpp/llama_grammar.py439CODE
LOWllama_cpp/llama_grammar.py500CODE
LOWllama_cpp/llama_grammar.py697CODE
LOWllama_cpp/llama_grammar.py446CODE
5 more matches not shown…
Unused Imports39 hits · 34 pts
SeverityFileLineSnippetContext
LOWtests/test_llama.py1CODE
LOWtests/test_llama.py6CODE
LOWexamples/low_level_api/ReasonAct.py2CODE
LOWexamples/server/server.py17CODE
LOWexamples/server/server.py39CODE
LOWexamples/batch-processing/server.py26CODE
LOWllama_cpp/llama_cache.py14CODE
LOWllama_cpp/llama_tokenizer.py1CODE
LOWllama_cpp/llama_chat_format.py1CODE
LOWllama_cpp/__init__.py1CODE
LOWllama_cpp/__init__.py2CODE
LOWllama_cpp/llama_cpp.py1CODE
LOWllama_cpp/llama_cpp.py16CODE
LOWllama_cpp/llama_cpp.py23CODE
LOWllama_cpp/llama_cpp.py23CODE
LOWllama_cpp/llama_cpp.py23CODE
LOWllama_cpp/_internals.py1CODE
LOWllama_cpp/_internals.py7CODE
LOWllama_cpp/_internals.py22CODE
LOWllama_cpp/llama.py1CODE
LOWllama_cpp/llama.py33CODE
LOWllama_cpp/llama.py35CODE
LOWllama_cpp/llama.py35CODE
LOWllama_cpp/llama.py35CODE
LOWllama_cpp/mtmd_cpp.py1CODE
LOWllama_cpp/mtmd_cpp.py5CODE
LOWllama_cpp/_ctypes_extensions.py1CODE
LOWllama_cpp/llama_cpp_ext.py7CODE
LOWllama_cpp/llava_cpp.py1CODE
LOWllama_cpp/llava_cpp.py4CODE
LOWllama_cpp/server/types.py1CODE
LOWllama_cpp/server/model.py1CODE
LOWllama_cpp/server/cli.py1CODE
LOWllama_cpp/server/settings.py1CODE
LOWllama_cpp/server/app.py1CODE
LOWllama_cpp/server/errors.py1CODE
LOWllama_cpp/server/errors.py6CODE
LOWllama_cpp/server/errors.py6CODE
LOWllama_cpp/server/__main__.py25CODE
Docstring Block Structure5 hits · 25 pts
SeverityFileLineSnippetContext
HIGHllama_cpp/llama.py123Load a llama.cpp model from `model_path`. Examples: Basic usage >>> import llama_cpp STRING
HIGHllama_cpp/llama.py605Tokenize a string. Args: text: The utf-8 encoded string to tokenize. add_bos: Whether tSTRING
HIGHllama_cpp/llama.py875Create a generator of tokens from a prompt. Examples: >>> llama = Llama("models/ggml-7b.bin") STRING
HIGHllama_cpp/llama.py1843Generate text from a prompt. Args: prompt: The prompt to generate text from. suffix: A STRING
HIGHllama_cpp/llama.py1940Generate text from a prompt. Args: prompt: The prompt to generate text from. suffix: A STRING
Self-Referential Comments8 hits · 22 pts
SeverityFileLineSnippetContext
MEDIUMdocker/open_llama/hug_model.py44 # Creating a symbolic link from destination to "model.bin"COMMENT
MEDIUMdocker/open_llama/hug_model.py74 # Create an argument parserCOMMENT
MEDIUMdocker/open_llama/hug_model.py92 # Define the parametersCOMMENT
MEDIUMscripts/releases-to-pep-503.sh33# Create an index html fileCOMMENT
MEDIUMscripts/releases-to-pep-503.sh49# Create an index html file in llama-cpp-python directoryCOMMENT
MEDIUMllama_cpp/llama_chat_format.py3160 """This method is no longer used in the new implementation."""STRING
MEDIUMllama_cpp/_internals.py637 # This method is deprecated in favor of using LlamaSampler directlyCOMMENT
MEDIUMllama_cpp/_internals.py876 # Create a new wrapper around the cloned samplerCOMMENT
Excessive Try-Catch Wrapping29 hits · 21 pts
SeverityFileLineSnippetContext
MEDIUMdocker/open_llama/hug_model.py135 print("Error - model choice was None")CODE
MEDIUMexamples/server/server.py3273def resolve_model_path(self) -> str:CODE
LOWexamples/server/server.py9349 except Exception:CODE
LOWexamples/server/server.py10831 except Exception:CODE
LOWexamples/server/server.py12488 except Exception:CODE
LOWexamples/server/server.py12507 except Exception: # noqa: BLE001CODE
LOWexamples/server/server.py13235 except Exception: # noqa: BLE001CODE
LOWexamples/server/server.py13309 except Exception: # noqa: BLE001CODE
LOWexamples/server/server.py13358 except Exception: # noqa: BLE001CODE
LOWexamples/server/server.py16256 except Exception:CODE
LOWexamples/server/server.py3194 except Exception:STRING
LOWexamples/server/server.py3232 except Exception as exc:STRING
LOWexamples/server/server.py3244 except Exception:STRING
LOWexamples/server/server.py3303 except Exception as exc:STRING
LOWllama_cpp/llama_chat_format.py717 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py1010 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py1737 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py2094 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py3085 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py3620 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py4315 except Exception as e:CODE
LOWllama_cpp/llama_chat_format.py4434 except Exception as e:CODE
LOWllama_cpp/llama.py494 except Exception as e:CODE
LOWllama_cpp/_ctypes_extensions.py93 except Exception as e:CODE
LOWllama_cpp/_ctypes_extensions.py103 except Exception as e:CODE
LOWllama_cpp/llama_grammar.py33 except Exception as err:CODE
LOWllama_cpp/server/errors.py180 except Exception as exc:CODE
LOWllama_cpp/server/errors.py198 except Exception:CODE
LOWllama_cpp/server/__main__.py80 except Exception as e:CODE
Over-Commented Block32 hits · 16 pts
SeverityFileLineSnippetContext
LOWexamples/server/server.py1#!/usr/bin/env -S uv run --scriptCOMMENT
LOWllama_cpp/llama_cpp.py61# GGML_TYPE_Q4_0 = 2,COMMENT
LOWllama_cpp/llama_cpp.py81# GGML_TYPE_IQ2_S = 22,COMMENT
LOWllama_cpp/llama_cpp.py201llama_token_p = ctypes.POINTER(llama_token)COMMENT
LOWllama_cpp/llama_cpp.py241# LLAMA_VOCAB_PRE_TYPE_DEEPSEEK_LLM = 2,COMMENT
LOWllama_cpp/llama_cpp.py261# LLAMA_VOCAB_PRE_TYPE_CODESHELL = 22,COMMENT
LOWllama_cpp/llama_cpp.py321# LLAMA_ROPE_TYPE_IMROPE = GGML_ROPE_TYPE_IMROPE,COMMENT
LOWllama_cpp/llama_cpp.py341LLAMA_TOKEN_TYPE_UNDEFINED = 0COMMENT
LOWllama_cpp/llama_cpp.py381# LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensorsCOMMENT
LOWllama_cpp/llama_cpp.py401# LLAMA_FTYPE_MOSTLY_IQ3_XXS = 23, // except 1d tensorsCOMMENT
LOWllama_cpp/llama_cpp.py481# LLAMA_POOLING_TYPE_NONE = 0,COMMENT
LOWllama_cpp/llama_cpp.py501LLAMA_ATTENTION_TYPE_NON_CAUSAL = 1COMMENT
LOWllama_cpp/llama_cpp.py621# typedef bool (*llama_progress_callback)(float progress, void * user_data);COMMENT
LOWllama_cpp/llama_cpp.py641# // )COMMENT
LOWllama_cpp/llama_cpp.py701COMMENT
LOWllama_cpp/llama_cpp.py821COMMENT
LOWllama_cpp/llama_cpp.py841# // Called with a progress value between 0.0 and 1.0. Pass NULL to disable.COMMENT
LOWllama_cpp/llama_cpp.py941# // https://github.com/ggml-org/llama.cpp/pull/7544COMMENT
LOWllama_cpp/llama_cpp.py961# float rope_freq_scale; // RoPE frequency scaling factor, 0 = from modelCOMMENT
LOWllama_cpp/llama_cpp.py981COMMENT
LOWllama_cpp/llama_cpp.py1141# int32_t nthread; // number of threads to use for quantizing, if <=0 will use std::thread::harCOMMENT
LOWllama_cpp/llama_cpp.py3141COMMENT
LOWllama_cpp/llama_cpp.py3241COMMENT
LOWllama_cpp/llama_cpp.py3501# // Wait until all computations are finishedCOMMENT
LOWllama_cpp/llama_cpp.py4281 _warn_deprecated("llama_vocab_cls", "use llama_vocab_bos instead")COMMENT
LOWllama_cpp/llama_cpp.py4341COMMENT
LOWllama_cpp/llama_cpp.py4441 text_len_max: The length of the buffer.COMMENT
LOWllama_cpp/llama_cpp.py4561# void (*apply) ( struct llama_sampler * smpl, llama_token_data_array * cur_p); // requiredCOMMENT
LOWllama_cpp/llama_cpp.py5221 """Set callback for all future logging events.COMMENT
LOWllama_cpp/mtmd_cpp.py241# float temp; // sampling temperature, or noise scale for flow-matching decodersCOMMENT
LOWllama_cpp/mtmd_cpp.py281 ("state_size", c_size_t),COMMENT
LOWllama_cpp/mtmd_cpp.py421COMMENT
Cross-Language Confusion2 hits · 12 pts
SeverityFileLineSnippetContext
HIGHtests/test_llama_chat_format.py75 "pad_token": null,CODE
HIGHllama_cpp/llama_cache.py138 # self.cache.push(_key, side="front") # type: ignoreCOMMENT
Redundant / Tautological Comments7 hits · 10 pts
SeverityFileLineSnippetContext
LOWscripts/get-releases.sh20 # Check if the response is valid JSONCOMMENT
LOWscripts/releases-to-pep-503.sh24# Check if all_releases.txt existsCOMMENT
LOW.github/workflows/build-wheels-cuda.yaml272 # Set tag_name to <tag>-cu<cuda_version>.COMMENT
LOWllama_cpp/llama_chat_format.py2816 # Check if vision is supportedCOMMENT
LOWllama_cpp/llama.py1475 # Check if stop sequence is in the tokenCOMMENT
LOWllama_cpp/server/app.py110 # Check if yaml fileCOMMENT
LOWllama_cpp/server/__main__.py63 # Check if yaml fileCOMMENT
AI Structural Patterns16 hits · 10 pts
SeverityFileLineSnippetContext
LOWexamples/server/server.py11239CODE
LOWllama_cpp/llama_chat_format.py1460CODE
LOWllama_cpp/llama_chat_format.py1823CODE
LOWllama_cpp/llama_chat_format.py4132CODE
LOWllama_cpp/llama_chat_format.py68CODE
LOWllama_cpp/llama_chat_format.py612CODE
LOWllama_cpp/llama_chat_format.py2850CODE
LOWllama_cpp/llama_chat_format.py3401CODE
LOWllama_cpp/llama.py60CODE
LOWllama_cpp/llama.py704CODE
LOWllama_cpp/llama.py792CODE
LOWllama_cpp/llama.py854CODE
LOWllama_cpp/llama.py1195CODE
LOWllama_cpp/llama.py1815CODE
LOWllama_cpp/llama.py1912CODE
LOWllama_cpp/llama.py2004CODE
Modern AI Meta-Vocabulary3 hits · 8 pts
SeverityFileLineSnippetContext
MEDIUMREADME.md310 # n_ctx=2048, # Uncomment to increase the context windowCOMMENT
MEDIUMREADME.md314 max_tokens=32, # Generate up to 32 tokens, set to None to generate up to the end of the context windowCODE
MEDIUMllama_cpp/llama.py1353 # Truncate max_tokens if requested tokens would exceed the context windowCOMMENT
Structural Annotation Overuse2 hits · 3 pts
SeverityFileLineSnippetContext
LOWllama_cpp/llama_types.py300# NOTE: The following type names are not part of the OpenAI OpenAPI specificationCOMMENT
LOWllama_cpp/llama_cpp.py4454# /// NOTE: This function does not use a jinja parser. It only support a pre-defined list of template. See more: https:/COMMENT
AI Slop Vocabulary1 hit · 3 pts
SeverityFileLineSnippetContext
MEDIUMllama_cpp/llama_cpp.py3088# // NOTE: this is a helper function to facilitate transition to the new batch API - avoid using itCOMMENT
TODO Padding1 hit · 2 pts
SeverityFileLineSnippetContext
LOWllama_cpp/llama_cpp.py3386# // TODO: implement get_embeddings() and get_logits() for llama_batch_extCOMMENT
Modern Structural Boilerplate4 hits · 1 pts
SeverityFileLineSnippetContext
LOWexamples/server/server.py12203 def set_draft_processing_enabled(self, enabled: bool) -> None:CODE
LOWexamples/server/server.py1201 def set_target_processing_enabled(self, enabled: bool) -> None:STRING
LOWexamples/server/server.py1398 def set_target_processing_enabled(self, enabled: bool) -> None:STRING
LOWexamples/server/server.py1461 def _set_nextn_layer_offset(self, offset: int) -> None:STRING