dphnAI/sonar

12.6

Adjusted Score

12.6

Raw Score

100%

Time Factor

2026-07-12

Last Push

1.8K

Stars

C++

Language

2.0M

Lines of Code

5.2K

Files

18.5K

Pattern Hits

2026-07-14

Scan Date

0.09

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 10HIGH 452MEDIUM 1951LOW 16099

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 18512 distinct pattern matches across 26 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

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

Hyper-Verbose Identifiers10648 hits · 10465 pts

Severity	File	Line	Snippet	Context
LOW	env.py	113	def run_and_parse_first_match(run_lambda, command, regex):	CODE
LOW	env.py	124	def run_and_return_first_line(run_lambda, command):	CODE
LOW	env.py	157	def get_nvidia_driver_version(run_lambda):	CODE
LOW	env.py	269	def summarize_aphrodite_build_flags():	CODE
LOW	env.py	473	def get_cachingallocator_config():	CODE
LOW	env.py	478	def get_cuda_module_loading_config():	CODE
LOW	setup.py	47	def should_require_rust_frontend() -> bool:	CODE
LOW	setup.py	52	def get_precompiled_rust_extension_paths() -> list[Path]:	CODE
LOW	setup.py	673	def add_aphrodite_package_data(filename: str) -> None:	CODE
LOW	csrc/cpu/generate_cpu_attn_dispatch.py	90	def generate_cases_for_isa_group(isa_list: list[str], include_fp8: bool = False) -> str:	CODE
LOW	csrc/libtorch_stable/quantization/machete/generate.py	318	def generate_type_option_name(kernel_types: TypeConfig):	STRING
LOW	csrc/libtorch_stable/quantization/machete/generate.py	350	def unsigned_type_with_bitwidth(num_bits):	STRING
LOW	tools/generate_versions_json.py	84	def generate_bake_native_json(args: dict[str, str]) -> dict:	CODE
LOW	tools/install_nixl_from_source_ubuntu.py	65	def install_system_dependencies():	CODE
LOW	tools/install_nixl_from_source_ubuntu.py	104	def build_and_install_prerequisites(args):	CODE
LOW	tools/build_rust.py	39	def rust_py_extension_module_names() -> list[str]:	CODE
LOW⚡	tools/pre_commit/generate_attention_backend_docs.py	876	def _expand_flash_attn_variants(	CODE
LOW⚡	tools/pre_commit/generate_attention_backend_docs.py	986	def parse_cuda_priority_lists() -> dict[str, list[str]]:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	249	def _find_exact_cc_in_function(tree: ast.AST, func_name: str) -> str \| None:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	607	def parse_flash_attn_features() -> dict[str, dict[str, Any]]:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	752	def parse_flashinfer_trtllm_features() -> dict[str, dict[str, Any]]:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	791	def parse_mla_prefill_backends() -> list[dict[str, Any]]:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	928	def _expand_flashinfer_variants(	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	1037	def _get_backends_from_return(stmts: list) -> list[str]:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	1307	def generate_priority_section(priorities: dict[str, list[str]]) -> str:	CODE
LOW	tools/profiler/visualize_layerwise_profile.py	66	def shorten_plot_legend_strings(legend, max_char_len: int):	CODE
LOW	tools/profiler/visualize_layerwise_profile.py	85	def attempt_to_make_names_unique(entries_and_traces):	CODE
LOW	tools/profiler/visualize_layerwise_profile.py	124	def group_trace_by_operations(trace_df: pd.DataFrame) -> pd.DataFrame:	CODE
LOW	tools/profiler/visualize_layerwise_profile.py	179	def is_custom_ar_all_reduce_unreg(op_name: str):	CODE
LOW	tests/test_sequence.py	9	def test_sequence_intermediate_tensors_equal():	CODE
LOW	tests/test_zen_cpu_platform_detection.py	35	def test_is_amd_zen_cpu_returns_false_when_cpuinfo_missing():	CODE
LOW	tests/test_version.py	36	def test_prev_minor_version_was(version_tuple, version_str, expected):	CODE
LOW⚡	tests/test_ray_env_utils.py	36	def test_arbitrary_var_propagated(self):	CODE
LOW⚡	tests/test_ray_env_utils.py	42	def test_worker_specific_excluded(self):	CODE
LOW⚡	tests/test_ray_env_utils.py	50	def test_non_carry_over_blacklist(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	20	def test_valueerror_converted_to_bypass(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	30	def test_original_valueerror_chained(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	44	def test_non_valueerror_propagates(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	54	def test_normal_return_preserved(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	76	def test_sentinel_attribute_set(self):	CODE
LOW	tests/test_fxgraphcache_pickle_patch.py	90	def test_patch_applied_in_current_environment():	CODE
LOW	tests/conftest.py	219	def init_test_http_connection():	CODE
LOW	tests/conftest.py	266	def should_do_global_cleanup_after_test(request) -> bool:	CODE
LOW	tests/conftest.py	720	def _hidden_states_to_seq_logprobs(	CODE
LOW	tests/conftest.py	743	def _hidden_states_to_logprobs(	CODE
LOW	tests/conftest.py	770	def generate_greedy_logprobs_limit(	CODE
LOW	tests/conftest.py	1031	def _final_steps_generate_w_logprobs(	CODE
LOW	tests/conftest.py	1139	def generate_prompt_perplexity(self, prompts: list[str], mask: Optional[list[str]] = None) -> list[float]:	CODE
LOW	tests/conftest.py	1244	def _wait_for_rocm_memory_release(self, gpu_memory_utilization: float) -> None:	CODE
LOW	tests/conftest.py	1284	def temporary_enable_log_propagate():	CODE
LOW	tests/conftest.py	1486	def pytest_collection_modifyitems(config, items):	CODE
LOW	tests/conftest.py	1503	def cli_config_file_with_model():	CODE
LOW	tests/conftest.py	1643	def clean_gpu_memory_between_tests():	CODE
LOW	tests/test_inputs.py	21	def test_preprocessor_always_mm_code_path(model_id, prompt):	CODE
LOW⚡	tests/test_ray_env.py	54	def test_pythonhashseed_in_result(self):	CODE
LOW⚡	tests/test_ray_env.py	118	def test_worker_specific_host_vars_are_excluded(self):	CODE
LOW⚡	tests/test_ray_env.py	128	def test_non_carry_over_blacklist(self):	CODE
LOW⚡	tests/test_ray_env.py	142	def test_additional_vars_passthrough(self):	CODE
LOW⚡	tests/test_ray_env.py	156	def test_prefix_no_partial_match(self):	CODE
LOW	tests/test_ray_env.py	168	def test_csv_whitespace_handling(self):	CODE
10588 more matches not shown…

Decorative Section Separators1542 hits · 5328 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	871	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	873	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	981	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	983	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	1424	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	1426	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	24	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	26	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	69	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	71	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	268	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	270	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	305	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	307	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	567	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	569	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	1107	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	1113	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	1199	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tools/pre_commit/generate_attention_backend_docs.py	1201	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_ray_env.py	11	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_ray_env.py	13	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	46	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	48	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	61	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	63	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_ray_env.py	97	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_ray_env.py	99	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	133	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	135	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	147	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_ray_env.py	149	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_jit_monitor.py	34	# ------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_jit_monitor.py	36	# ------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_jit_monitor.py	121	# ------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_jit_monitor.py	123	# ------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_jit_monitor.py	434	# ------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_jit_monitor.py	436	# ------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	123	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	125	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	147	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	149	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	167	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	169	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	187	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	189	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	208	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	210	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	240	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_factory.py	242	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/v1/kv_offload/test_factory.py	29	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/v1/kv_offload/test_factory.py	31	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/v1/kv_offload/test_file_mapper.py	21	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/v1/kv_offload/test_file_mapper.py	23	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	44	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	46	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	79	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	81	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	132	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	134	# ---------------------------------------------------------------------------	COMMENT
1482 more matches not shown…

Cross-File Repetition358 hits · 1790 pts

Severity	File	Line	Snippet	Context
HIGH	tests/v1/attention/test_mla_prefill_selector.py	0	clear lru cache to ensure each test case runs without caching.	STRING
HIGH	tests/kernels/attention/test_attention_selector.py	0	clear lru cache to ensure each test case runs without caching.	STRING
HIGH	tests/kernels/attention/test_mha_attn.py	0	clear lru cache to ensure each test case runs without caching.	STRING
HIGH	tests/kernels/attention/test_rocm_attention_selector.py	0	clear lru cache to ensure each test case runs without caching.	STRING
HIGH	tests/tool_parsers/test_granite_tool_parser.py	0	<longcat_tool_call>{ "name": "test_function", "arguments": { "string_field": "hello", "int_field": 42, "float_field": 3.	STRING
HIGH	tests/tool_parsers/test_granite_20b_fc_tool_parser.py	0	<longcat_tool_call>{ "name": "test_function", "arguments": { "string_field": "hello", "int_field": 42, "float_field": 3.	STRING
HIGH	tests/tool_parsers/test_phi4mini_tool_parser.py	0	<longcat_tool_call>{ "name": "test_function", "arguments": { "string_field": "hello", "int_field": 42, "float_field": 3.	STRING
HIGH	tests/tool_parsers/test_longcat_tool_parser.py	0	<longcat_tool_call>{ "name": "test_function", "arguments": { "string_field": "hello", "int_field": 42, "float_field": 3.	STRING
HIGH	tests/kernels/quantization/test_block_int8.py	0	sets the default cuda device for all tests in this module.	STRING
HIGH	tests/kernels/quantization/test_int8_kernel.py	0	sets the default cuda device for all tests in this module.	STRING
HIGH	tests/kernels/moe/test_block_int8.py	0	sets the default cuda device for all tests in this module.	STRING
HIGH	tests/kernels/moe/test_triton_moe_ptpc_fp8.py	0	sets the default cuda device for all tests in this module.	STRING
HIGH	tests/distributed/test_pipeline_parallel.py	0	warning: this test runs in both single-node (4 gpus) and multi-node (2 node with 2 gpus each) modes. if the test only us	STRING
HIGH	tests/distributed/test_context_parallel.py	0	warning: this test runs in both single-node (4 gpus) and multi-node (2 node with 2 gpus each) modes. if the test only us	STRING
HIGH	tests/compile/correctness_e2e/test_sequence_parallel.py	0	warning: this test runs in both single-node (4 gpus) and multi-node (2 node with 2 gpus each) modes. if the test only us	STRING
HIGH	tests/models/multimodal/pooling/test_colpali.py	0	create a small solid-color png image and return its base64 data uri.	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3.py	0	create a small solid-color png image and return its base64 data uri.	STRING
HIGH	tests/entrypoints/pooling/scoring/util.py	0	create a small solid-color png image and return its base64 data uri.	STRING
HIGH	tests/models/multimodal/pooling/test_colpali.py	0	build a scoremultimodalparam containing an image (and optional text).	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3.py	0	build a scoremultimodalparam containing an image (and optional text).	STRING
HIGH	tests/entrypoints/pooling/scoring/util.py	0	build a scoremultimodalparam containing an image (and optional text).	STRING
HIGH	tests/models/multimodal/pooling/test_colpali.py	0	verify per-token embedding shape and l2 normalization.	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3.py	0	verify per-token embedding shape and l2 normalization.	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3_5.py	0	verify per-token embedding shape and l2 normalization.	STRING
HIGH	tests/models/multimodal/pooling/test_colpali.py	0	verify that relevant documents score higher than irrelevant ones.	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3.py	0	verify that relevant documents score higher than irrelevant ones.	STRING
HIGH	tests/models/multimodal/pooling/test_colqwen3_5.py	0	verify that relevant documents score higher than irrelevant ones.	STRING
HIGH	tests/models/multimodal/generation/test_phi4siglip.py	0	sanitize aphrodite output [phi3v] to be comparable with hf output.	STRING
HIGH	tests/models/multimodal/generation/test_phi4mm.py	0	sanitize aphrodite output [phi3v] to be comparable with hf output.	STRING
HIGH	…/models/multimodal/generation/vlm_utils/model_utils.py	0	sanitize aphrodite output [phi3v] to be comparable with hf output.	STRING
HIGH	…ts/models/multimodal/generation/test_granite_speech.py	0	sanitize aphrodite output [phi3v] to be comparable with hf output.	STRING
HIGH	tests/entrypoints/openai/completion/test_completion.py	0	root ::= select_statement select_statement ::= "select " column " from " table " where " condition column ::= "col_1 " \|	STRING
HIGH	…ypoints/openai/chat_completion/test_chat_completion.py	0	root ::= select_statement select_statement ::= "select " column " from " table " where " condition column ::= "col_1 " \|	STRING
HIGH	benchmarks/benchmark_serving_structured_output.py	0	root ::= select_statement select_statement ::= "select " column " from " table " where " condition column ::= "col_1 " \|	STRING
HIGH	tests/lora/test_qwen3moe_tp.py	0	i want you to act as a sql terminal in front of an example database, you need only to return the sql command to me. do n	STRING
HIGH	tests/lora/test_llama_tp.py	0	i want you to act as a sql terminal in front of an example database, you need only to return the sql command to me. do n	STRING
HIGH	tests/lora/test_olmoe_tp.py	0	i want you to act as a sql terminal in front of an example database, you need only to return the sql command to me. do n	STRING
HIGH	examples/offline_inference/vision_language_embedding.py	0	this example shows how to use aphrodite for running offline inference with the correct prompt format on audio language m	STRING
HIGH	examples/vision/vision_example.py	0	this example shows how to use aphrodite for running offline inference with the correct prompt format on audio language m	STRING
HIGH	examples/audio/audio_example.py	0	this example shows how to use aphrodite for running offline inference with the correct prompt format on audio language m	STRING
HIGH	benchmarks/benchmark_throughput.py	0	deprecated: this script has been moved to the aphrodite cli. please use the following command instead: aphrodite bench r	STRING
HIGH	benchmarks/benchmark_latency.py	0	deprecated: this script has been moved to the aphrodite cli. please use the following command instead: aphrodite bench r	STRING
HIGH	benchmarks/benchmark_serving.py	0	deprecated: this script has been moved to the aphrodite cli. please use the following command instead: aphrodite bench r	STRING
HIGH	benchmarks/kernels/benchmark_machete.py	0	benchmark cutlass gemm. to run square gemms: python3 ./benchmarks/cutlass_benchmarks/sparse_benchmarks.py --dtype fp8 sq	STRING
HIGH	benchmarks/cutlass_benchmarks/w8a8_benchmarks.py	0	benchmark cutlass gemm. to run square gemms: python3 ./benchmarks/cutlass_benchmarks/sparse_benchmarks.py --dtype fp8 sq	STRING
HIGH	benchmarks/cutlass_benchmarks/sparse_benchmarks.py	0	benchmark cutlass gemm. to run square gemms: python3 ./benchmarks/cutlass_benchmarks/sparse_benchmarks.py --dtype fp8 sq	STRING
HIGH	benchmarks/kernels/benchmark_w8a8_block_fp8.py	0	this function performs matrix multiplication with block-wise quantization. it takes two input tensors `a` and `b` with s	STRING
HIGH	…model_executor/layers/quantization/utils/int8_utils.py	0	this function performs matrix multiplication with block-wise quantization. it takes two input tensors `a` and `b` with s	STRING
HIGH	…/model_executor/layers/quantization/utils/fp8_utils.py	0	this function performs matrix multiplication with block-wise quantization. it takes two input tensors `a` and `b` with s	STRING
HIGH	aphrodite/sampling_params.py	0	whether to include the stop strings in output text.	STRING
HIGH	…e/entrypoints/speech_to_text/transcription/protocol.py	0	whether to include the stop strings in output text.	STRING
HIGH	…ite/entrypoints/speech_to_text/translation/protocol.py	0	whether to include the stop strings in output text.	STRING
HIGH	aphrodite/_aiter_ops.py	0	forward pass for mhc pre block. args: residual: shape (..., hc_mult, hidden_size), dtype torch.bfloat16 fn: shape (hc_mu	STRING
HIGH	aphrodite/model_executor/kernels/mhc/aiter.py	0	forward pass for mhc pre block. args: residual: shape (..., hc_mult, hidden_size), dtype torch.bfloat16 fn: shape (hc_mu	STRING
HIGH	aphrodite/model_executor/kernels/mhc/torch.py	0	forward pass for mhc pre block. args: residual: shape (..., hc_mult, hidden_size), dtype torch.bfloat16 fn: shape (hc_mu	STRING
HIGH	aphrodite/v1/attention/backends/flash_attn.py	0	forward pass with aiterflashattention. args: query: shape = [num_tokens, num_heads, head_size] key: shape = [num_tokens,	STRING
HIGH	aphrodite/v1/attention/backends/flash_attn_diffkv.py	0	forward pass with aiterflashattention. args: query: shape = [num_tokens, num_heads, head_size] key: shape = [num_tokens,	STRING
HIGH	aphrodite/v1/attention/backends/rocm_aiter_fa.py	0	forward pass with aiterflashattention. args: query: shape = [num_tokens, num_heads, head_size] key: shape = [num_tokens,	STRING
HIGH	aphrodite/v1/attention/backends/flash_attn.py	0	forward pass for encoder attention without kv cache. args: query: shape = [num_encoder_tokens, num_heads, head_size] key	STRING
HIGH	aphrodite/v1/attention/backends/triton_attn.py	0	forward pass for encoder attention without kv cache. args: query: shape = [num_encoder_tokens, num_heads, head_size] key	STRING
298 more matches not shown…

Unused Imports1436 hits · 1334 pts

Severity	File	Line	Context
LOW	…lass_extensions/aphrodite_cutlass_library_extension.py	6	CODE
LOW	tools/build_rust.py	6	CODE
LOW	tests/conftest.py	77	CODE
LOW	tests/conftest.py	77	CODE
LOW	tests/conftest.py	78	CODE
LOW	tests/utils.py	62	CODE
LOW	tests/utils.py	2049	CODE
LOW	tests/v1/kv_offload/tiering/p2p/test_sessions.py	13	CODE
LOW	tests/v1/kv_offload/tiering/p2p/test_zmq_transport.py	5	CODE
LOW	tests/v1/kv_offload/tiering/p2p/test_data_transport.py	5	CODE
LOW	tests/v1/kv_offload/tiering/p2p/test_manager.py	9	CODE
LOW	tests/v1/attention/test_mla_backends.py	71	CODE
LOW	tests/v1/attention/test_attention_backends.py	51	CODE
LOW	tests/v1/logits_processors/utils.py	12	CODE
LOW	tests/v1/cudagraph/test_cudagraph_mode.py	38	CODE
LOW	tests/v1/cudagraph/test_breakable_cudagraph.py	7	CODE
LOW	tests/v1/kv_connector/unit/test_tp_mapping.py	10	CODE
LOW	tests/v1/kv_connector/unit/test_hf3fs_client.py	17	CODE
LOW	tests/v1/kv_connector/unit/test_hf3fs_client.py	17	CODE
LOW	tests/v1/kv_connector/unit/test_hf3fs_client.py	17	CODE
LOW	tests/v1/kv_connector/unit/test_hf3fs_client.py	17	CODE
LOW	tests/v1/kv_connector/unit/test_hf3fs_client.py	17	CODE
LOW	tests/v1/kv_connector/unit/test_nixl_push_connector.py	21	CODE
LOW	…/v1/kv_connector/unit/offloading_connector/conftest.py	3	CODE
LOW	tests/v1/spec_decode/test_backup_token_async_spec.py	9	CODE
LOW	tests/v1/sample/test_topk_topp_sampler.py	33	CODE
LOW	tests/v1/engine/conftest.py	20	CODE
LOW	tests/v1/engine/conftest.py	20	CODE
LOW	tests/v1/simple_kv_offload/test_scheduler.py	5	CODE
LOW	tests/v1/simple_kv_offload/test_worker.py	10	CODE
LOW	tests/tool_use/test_gemma4_responses_adjust_request.py	32	CODE
LOW	tests/tool_parsers/test_poolside_v1_tool_parser.py	21	CODE
LOW	tests/renderers/test_chat_utils_prompt_embeds.py	6	CODE
LOW	tests/kernels/core/test_vit_fp8_attn.py	20	CODE
LOW	tests/kernels/core/test_fused_q_kv_rmsnorm.py	11	CODE
LOW	tests/kernels/ir/test_ir_ops.py	11	CODE
LOW	tests/kernels/ir/test_layernorm.py	7	CODE
LOW	tests/kernels/mamba/test_precopy_mamba_align.py	20	CODE
LOW	tests/kernels/mamba/test_gdn_forward_core_split.py	30	CODE
LOW	tests/kernels/mamba/test_ssu_dispatch.py	24	CODE
LOW	tests/kernels/moe/test_moe.py	18	CODE
LOW	tests/distributed/test_eplb_spec_decode.py	3	CODE
LOW	…ummy_platform/aphrodite_add_dummy_platform/__init__.py	10	CODE
LOW	tests/cuda/scripts/check_device_count_respects_env.py	14	CODE
LOW	tests/model_executor/test_oink_integration.py	37	CODE
LOW	…/aphrodite_test_utils/aphrodite_test_utils/__init__.py	8	CODE
LOW	…/aphrodite_test_utils/aphrodite_test_utils/__init__.py	8	CODE
LOW	…/aphrodite_test_utils/aphrodite_test_utils/__init__.py	9	CODE
LOW	…/aphrodite_test_utils/aphrodite_test_utils/__init__.py	9	CODE
LOW	tests/models/language/pooling/embed_utils.py	8	CODE
LOW	tests/models/language/pooling/test_reward.py	18	CODE
LOW	tests/models/multimodal/generation/test_pixtral.py	24	CODE
LOW	tests/parser/engine/test_replay.py	13	CODE
LOW	tests/parser/engine/trace_builder.py	14	CODE
LOW	tests/parser/engine/test_delegating_replay.py	14	CODE
LOW	tests/parser/engine/streaming_helpers.py	5	CODE
LOW	tests/parser/engine/replay_harness.py	10	CODE
LOW	tests/parser/engine/test_parser_engine.py	8	CODE
LOW	tests/parser/engine/test_ufffd_reasoning_transition.py	14	CODE
LOW	tests/reasoning/test_cohere_command_reasoning_parser.py	4	CODE
1376 more matches not shown…

Deep Nesting1110 hits · 1003 pts

Severity	File	Line	Context
LOW	env.py	165	CODE
LOW	env.py	379	CODE
LOW	setup.py	495	CODE
LOW	setup.py	538	CODE
LOW	setup.py	173	CODE
LOW	csrc/cpu/generate_cpu_attn_dispatch.py	90	CODE
LOW	…ibtorch_stable/quantization/marlin/generate_kernels.py	173	CODE
LOW	…btorch_stable/moe/marlin_moe_wna16/generate_kernels.py	173	CODE
LOW	tools/report_build_time_ninja.py	149	CODE
LOW	tools/pre_commit/check_triton_import.py	32	CODE
LOW	tools/pre_commit/enforce_regex_import.py	34	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	105	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	130	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	151	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	194	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	319	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	361	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	607	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	986	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	1037	CODE
LOW	tools/pre_commit/validate_config.py	73	CODE
LOW	tools/pre_commit/check_boolean_context_manager.py	8	CODE
LOW	tools/pre_commit/check_spdx_header.py	23	CODE
LOW	tools/pre_commit/check_spdx_header.py	49	CODE
LOW	tests/conftest.py	523	CODE
LOW	tests/conftest.py	954	CODE
LOW	tests/utils.py	1141	CODE
LOW	tests/utils.py	1461	CODE
LOW	tests/utils.py	1588	CODE
LOW	tests/utils.py	1712	CODE
LOW	tests/utils.py	538	CODE
LOW	tests/utils.py	1594	CODE
LOW	tests/utils.py	1735	CODE
LOW	tests/v1/utils.py	12	CODE
LOW	tests/v1/kv_offload/cpu/test_gpu_worker.py	216	CODE
LOW	tests/v1/tracing/test_tracing.py	22	CODE
LOW	tests/v1/attention/test_mla_backends.py	877	CODE
LOW	tests/v1/attention/test_sparse_mla_backends.py	493	CODE
LOW	tests/v1/logits_processors/test_correctness.py	289	CODE
LOW	tests/v1/logits_processors/test_correctness.py	433	CODE
LOW	tests/v1/core/test_scheduler.py	2234	CODE
LOW	tests/v1/core/utils.py	196	CODE
LOW	tests/v1/kv_connector/unit/test_nixl_connector.py	2742	CODE
LOW	tests/v1/kv_connector/unit/test_offloading_connector.py	301	CODE
LOW	tests/v1/kv_connector/unit/test_offloading_connector.py	353	CODE
LOW	tests/v1/kv_connector/unit/test_offloading_connector.py	365	CODE
LOW	tests/v1/kv_connector/unit/utils.py	299	CODE
LOW	tests/v1/kv_connector/unit/test_mooncake_connector.py	1162	CODE
LOW	…kv_connector/unit/offloading_connector/test_metrics.py	217	CODE
LOW	tests/v1/determinism/test_batch_invariance.py	28	CODE
LOW	tests/v1/determinism/test_batch_invariance.py	154	CODE
LOW	tests/v1/determinism/test_batch_invariance.py	610	CODE
LOW	tests/v1/determinism/test_nvfp4_batch_invariant.py	41	CODE
LOW	tests/v1/spec_decode/test_eagle.py	48	CODE
LOW	tests/v1/spec_decode/test_acceptance_length.py	173	CODE
LOW	tests/v1/spec_decode/test_acceptance_length.py	218	CODE
LOW	tests/v1/sample/test_logprobs.py	139	CODE
LOW	tests/v1/sample/test_logprobs.py	496	CODE
LOW	tests/v1/sample/test_logprobs.py	868	CODE
LOW	tests/v1/sample/test_logprobs.py	909	CODE
1050 more matches not shown…

Self-Referential Comments289 hits · 926 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	build_and_upload_docker.sh	31	# Create a temporary container to extract the wheel	COMMENT
MEDIUM	tools/install_deepgemm.sh	84	# Create a temporary directory for the build	COMMENT
MEDIUM	tools/report_build_time_ninja.py	199	# Create a list that is in order by time stamp and has entries for the	COMMENT
MEDIUM	tests/conftest.py	549	# Create a copy to avoid modifying the original dict	COMMENT
MEDIUM	tests/utils.py	733	# Create a dedicated process group so we can kill	COMMENT
MEDIUM	tests/utils.py	1597	# Create a unique temporary file to store exception info from child	COMMENT
MEDIUM⚡	tests/test_access_log_filter.py	257	# Create a logger with our filter (simulating uvicorn.access)	COMMENT
MEDIUM⚡	tests/test_access_log_filter.py	264	# Create a custom handler that tracks messages	COMMENT
MEDIUM	tests/test_config.py	812	# Create a new mock and run the method with the same S3 URL	COMMENT
MEDIUM	tests/v1/test_tensor_ipc_queue.py	185	# Create a CPU tensor	COMMENT
MEDIUM	tests/v1/test_tensor_ipc_queue.py	491	# Create a CPU tensor	COMMENT
MEDIUM	tests/v1/test_tensor_ipc_queue.py	616	# Create a CPU tensor	COMMENT
MEDIUM	tests/v1/test_tensor_ipc_queue.py	859	# Create a tensor queue	COMMENT
MEDIUM	tests/v1/test_serial_utils.py	185	# Create a sample Python object	COMMENT
MEDIUM	tests/v1/test_serial_utils.py	203	# Create a sample tensor	COMMENT
MEDIUM	tests/v1/test_serial_utils.py	221	# Create a sample numpy array	COMMENT
MEDIUM	tests/v1/test_serial_utils.py	305	# Create a request with a non-multimodal tensor	COMMENT
MEDIUM	tests/v1/test_serial_utils.py	344	# Create a request with None for the tensor field	COMMENT
MEDIUM⚡	tests/v1/kv_offload/test_file_mapper.py	51	# Create a copy of the mock config to avoid modifying the global one	COMMENT
MEDIUM	tests/v1/metrics/test_ray_metrics.py	67	# Create the actor and call the async method	COMMENT
MEDIUM	tests/v1/attention/test_mla_backends.py	258	# Create a realistic slot mapping that corresponds to the block table	COMMENT
MEDIUM	tests/v1/attention/test_mla_backends.py	1306	# Create a summary for the single-line failure message	COMMENT
MEDIUM	tests/v1/attention/test_attention_backends.py	172	# Create a realistic slot mapping that corresponds to the block table	COMMENT
MEDIUM	tests/v1/logits_processors/test_correctness.py	743	# Define a shuffled batch of requests which individually use a different	COMMENT
MEDIUM	tests/v1/logits_processors/test_custom_offline.py	28	# Create a mixture of requests which do and don't utilize the dummy logitproc	COMMENT
MEDIUM	tests/v1/logits_processors/test_custom_offline.py	63	# Create a Aphrodite instance and load custom logitproc	COMMENT
MEDIUM	tests/v1/logits_processors/test_custom_offline.py	70	# Create a reference Aphrodite instance without custom logitproc	COMMENT
MEDIUM⚡	tests/v1/core/test_kv_cache_utils.py	257	# Create a list of KVCacheBlock objects	COMMENT
MEDIUM⚡	tests/v1/core/test_kv_cache_utils.py	260	# Create a FreeKVCacheBlockQueue with these blocks	COMMENT
MEDIUM⚡	tests/v1/core/test_kv_cache_utils.py	457	# Create a list of KVCacheBlock objects	COMMENT
MEDIUM⚡	tests/v1/core/test_kv_cache_utils.py	460	# Create a FreeKVCacheBlockQueue with these blocks	COMMENT
MEDIUM	tests/v1/core/test_kv_cache_utils.py	303	# Create an empty FreeKVCacheBlockQueue with these blocks	COMMENT
MEDIUM	tests/v1/core/test_kv_cache_utils.py	351	# Create an empty FreeKVCacheBlockQueue	COMMENT
MEDIUM	tests/v1/core/test_kv_cache_utils.py	402	# Create an empty FreeKVCacheBlockQueue with these blocks	COMMENT
MEDIUM	tests/v1/core/test_kv_cache_utils.py	1292	# Create a AphroditeConfig	COMMENT
MEDIUM	tests/v1/core/test_kv_cache_utils.py	1326	# Create a AphroditeConfig	COMMENT
MEDIUM⚡	tests/v1/core/test_scheduler.py	2927	# Create a request and schedule it	COMMENT
MEDIUM⚡	tests/v1/core/test_scheduler.py	3685	# Create a request and schedule it (and to be preempted)	COMMENT
MEDIUM	tests/v1/core/test_scheduler.py	2954	# Create a high priority request and schedule it	COMMENT
MEDIUM	tests/v1/core/test_scheduler.py	3733	# Create a high priority request and schedule it	COMMENT
MEDIUM	tests/v1/core/test_scheduler.py	4556	# Create a text-only request (no mm_features).	COMMENT
MEDIUM	tests/v1/cudagraph/test_cudagraph_dispatch.py	56	# Create a real LoRAConfig with specialize_active_lora enabled	COMMENT
MEDIUM	tests/v1/kv_connector/unit/test_nixl_connector.py	2124	# Create a request that triggers do_remote_decode so that	COMMENT
MEDIUM⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	202	# Create a mock object that is not LMCacheKVEvents	COMMENT
MEDIUM	tests/v1/kv_connector/unit/test_moriio_connector.py	209	# Define a fake remote engine id for testing	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	139	# Create a request with multiple blocks worth of tokens	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	183	# Create a request	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	205	# Create a request with just 1 token	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	223	# Create a request with 2 tokens	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	249	# Create a request with many blocks	COMMENT
MEDIUM	…ts/v1/kv_connector/unit/test_decode_bench_connector.py	332	# Create a request that doesn't align to block boundaries	COMMENT
MEDIUM	tests/v1/kv_connector/unit/test_nixl_connector_hma.py	73	# Create a mock worker with just the required attributes	COMMENT
MEDIUM	tests/v1/kv_connector/unit/test_example_connector.py	144	# Create the LLM instance	COMMENT
MEDIUM	…r/extract_hidden_states_integration/test_extraction.py	45	# Create a minimal Llama config with small dimensions	COMMENT
MEDIUM	…r/extract_hidden_states_integration/test_extraction.py	60	# Create a simple tokenizer	COMMENT
MEDIUM	tests/v1/determinism/test_batch_invariance.py	103	# Create a batch of size `max_batch_size` and insert the needle at	COMMENT
MEDIUM	tests/v1/distributed/test_external_lb_dp.py	149	# Create a client for each server	COMMENT
MEDIUM	tests/v1/distributed/test_hybrid_lb_dp.py	182	# Create a client for each node (each node has its own API endpoint)	COMMENT
MEDIUM	tests/v1/streaming_input/test_async_llm_streaming.py	20	# Create a minimal mock without initializing the full engine	COMMENT
MEDIUM	tests/v1/streaming_input/test_async_llm_streaming.py	50	# Create a mock queue with outputs	COMMENT
229 more matches not shown…

Over-Commented Block969 hits · 847 pts

Severity	File	Line	Snippet	Context
LOW	CMakeLists.txt	1	cmake_minimum_required(VERSION 3.26)	COMMENT
LOW	CMakeLists.txt	61	#	COMMENT
LOW	CMakeLists.txt	101	find_program(NVCC_EXECUTABLE nvcc)	COMMENT
LOW	CMakeLists.txt	241	endif()	COMMENT
LOW	env.py	281	return None	COMMENT
LOW	env.py	301	# Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr	COMMENT
LOW	env.py	321	# NUMA node1 CPU(s): 32-63,96-127	COMMENT
LOW	env.py	341	# Manufacturer=GenuineIntel	COMMENT
LOW	config.yaml	1	# Sample configuration file for Aphrodite Engine	COMMENT
LOW	config.yaml	21		COMMENT
LOW	config.yaml	41	# at high contexts, but it mutually exclusive with kv_cache_dtype=fp8	COMMENT
LOW	config.yaml	81	# The data type to use for the model. Can be `auto`, `float16`, `bfloat16`,	COMMENT
LOW	config.yaml	101	# Otherwise, KV cache scaling factors default to 1.0, which	COMMENT
LOW	config.yaml	121		COMMENT
LOW	config.yaml	141	speculative_args:	COMMENT
LOW	config.yaml	161		COMMENT
LOW	config.yaml	181	# of draft tokens. Less accurate but faster.	COMMENT
LOW	config.yaml	241	- max_cpu_loras:	COMMENT
LOW	config.yaml	301		COMMENT
LOW	config.yaml	321	# to the chosen load_format. This should be a JSON string that	COMMENT
LOW	formatting.sh	1	#!/usr/bin/env bash	COMMENT
LOW	csrc/fork_torch_bindings.cpp	1	// Fork-specific op registrations (extension: _C_fork).	COMMENT
LOW	csrc/torch_utils.h	1	#pragma once	COMMENT
LOW	csrc/torch_bindings.cpp	1	// Provides torch::Tensor for ops.h (previously included transitively via	COMMENT
LOW	csrc/cumem_allocator_compat.h	101	} // extern "C"	COMMENT
LOW	csrc/cumem_allocator.cpp	1	// A CUDAPluggableAllocator based on cumem* APIs.	COMMENT
LOW	csrc/cuda_compat.h	41	#define APHRODITE_LDG(arg) __ldg(arg)	COMMENT
LOW	csrc/cuda_compat.h	61	#define APHRODITE_SHFL_SYNC(var, src_lane) __shfl(var, src_lane)	COMMENT
LOW	csrc/cuda_utils.h	1	#pragma once	COMMENT
LOW	csrc/spinloop.cpp	1	#include <Python.h>	COMMENT
LOW	csrc/attention/attention_dtypes.h	1	#pragma once	COMMENT
LOW	csrc/hadamard/static_switch.h	1	// Inspired by	COMMENT
LOW	csrc/core/scalar_type.hpp	1	#pragma once	COMMENT
LOW	csrc/core/registration.h	1	#pragma once	COMMENT
LOW	csrc/metal/paged_ops.cpp	1	// SPDX-License-Identifier: Apache-2.0	COMMENT
LOW	csrc/metal/paged_ops.cpp	741	// Replaces the Python turbo_quant_encode() + 5 MLX scatters on the hot path.	COMMENT
LOW	csrc/cpu/cpu_attn_neon.hpp	1	#ifndef CPU_ATTN_NEON_HPP	COMMENT
LOW	csrc/cpu/utils.cpp	1	#ifndef APHRODITE_NUMA_DISABLED	COMMENT
LOW	csrc/cpu/cpu_wna16.cpp	1	#include "cpu/cpu_types.hpp"	COMMENT
LOW	csrc/cpu/cpu_fused_moe.cpp	1	#include "cpu/cpu_types.hpp"	COMMENT
LOW	csrc/cpu/cpu_types.hpp	1	#ifndef CPU_TYPES_HPP	COMMENT
LOW	csrc/cpu/cpu_types.hpp	21	#include "cpu_types_scalar.hpp"	COMMENT
LOW	csrc/cpu/cpu_types_riscv.hpp	1	#ifndef CPU_TYPES_RISCV_HPP	COMMENT
LOW	csrc/cpu/cpu_attn_impl.hpp	981	// - q_heads_buffer: [MaxQHeadNumPerIteration, head_dim]	COMMENT
LOW	csrc/cpu/cpu_attn_impl.hpp	1001	// - q_heads_per_kv	COMMENT
LOW	csrc/cpu/cpu_attn_impl.hpp	1121	// print_logits("masked logits", logits_buffer, q_head_num,	COMMENT
LOW	csrc/cpu/cpu_attn_impl.hpp	1821		COMMENT
LOW	csrc/cpu/cpu_attn_impl.hpp	1841		COMMENT
LOW	csrc/cpu/cpu_attn_vxe.hpp	381	} // namespace cpu_attention	COMMENT
LOW	csrc/cpu/generate_cpu_attn_dispatch.py	141	#endif	COMMENT
LOW	csrc/cpu/cpu_arch_macros.h	61	#endif	COMMENT
LOW	csrc/cpu/cpu_arch_macros.h	161	#include <riscv_vector.h>	COMMENT
LOW	csrc/cpu/cpu_attn_rvv.hpp	1	// SPDX-License-Identifier: Apache-2.0	COMMENT
LOW	csrc/cpu/cpu_attn_fp8.hpp	1	// SPDX-License-Identifier: Apache-2.0	COMMENT
LOW	csrc/cpu/cpu_types_vxe.hpp	1		COMMENT
LOW	csrc/cpu/cpu_types_vxe.hpp	21	#define vec_sr(a, b) ((a) >> (b)) // Vector Shift Right Algebraic	COMMENT
LOW	csrc/cpu/cpu_types_vxe.hpp	1201	#endif	COMMENT
LOW	csrc/cpu/cpu_types_x86.hpp	1		COMMENT
LOW	csrc/cpu/shm.cpp	1	#include "cpu/cpu_types.hpp"	COMMENT
LOW	csrc/cpu/cpu_types_riscv_impl.hpp	1	#ifndef CPU_TYPES_RISCV_IMPL_HPP	COMMENT
909 more matches not shown…

Excessive Try-Catch Wrapping629 hits · 669 pts

Severity	File	Line	Snippet	Context
LOW	env.py	256	except Exception:	CODE
LOW	setup.py	115	except Exception:	CODE
LOW	setup.py	207	except Exception as e:	CODE
LOW	setup.py	478	except Exception:	CODE
MEDIUM	setup.py	107	def find_tcmalloc() -> Path \| None:	CODE
MEDIUM	tools/generate_cmake_presets.py	156	print(f"Error writing file: {e}")	CODE
LOW	tools/install_nixl_from_source_ubuntu.py	30	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	501	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	515	except Exception as e:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	583	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	618	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	764	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	805	except Exception:	CODE
LOW	tools/pre_commit/generate_attention_backend_docs.py	1007	except Exception:	CODE
LOW	tests/conftest.py	1269	except Exception:	CODE
LOW	tests/conftest.py	1530	except Exception as e:	CODE
MEDIUM	tests/utils.py	117	def _nvml():	CODE
LOW	tests/utils.py	81	except Exception as e:	CODE
LOW	tests/utils.py	292	except Exception:	CODE
LOW	tests/utils.py	574	except Exception as e:	CODE
LOW	tests/utils.py	654	except Exception:	CODE
LOW	tests/utils.py	1626	except Exception as e:	CODE
LOW	tests/utils.py	1649	except Exception:	CODE
LOW	tests/utils.py	1840	except Exception as e:	CODE
LOW	tests/utils.py	2052	except Exception:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	81	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	116	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	300	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	397	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	429	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	512	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	548	except Exception as e:	CODE
LOW	tests/v1/test_tensor_ipc_queue.py	691	except Exception as e:	CODE
LOW	tests/v1/utils.py	62	except Exception as e:	CODE
LOW⚡	tests/v1/kv_offload/cpu/test_shared_offload_region.py	156	except Exception as e:	CODE
LOW	tests/v1/kv_offload/cpu/test_shared_offload_region.py	117	except Exception as e:	CODE
LOW	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	214	except Exception as e:	CODE
LOW	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	253	except Exception as exc:	CODE
LOW	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	264	except Exception as exc:	CODE
LOW	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	284	except Exception as e:	CODE
MEDIUM	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	249	def _run_decode():	CODE
LOW	tests/v1/shutdown/test_forward_error.py	73	except Exception as e:	CODE
LOW	tests/v1/shutdown/test_processor_error.py	37	except Exception as e:	CODE
LOW⚡	tests/v1/kv_connector/unit/test_hf3fs_client.py	28	except Exception:	CODE
LOW	tests/v1/kv_connector/unit/test_multi_connector.py	373	except Exception as e:	CODE
LOW	tests/v1/kv_connector/unit/utils.py	334	except Exception as e:	CODE
LOW	…/v1/kv_connector/unit/test_mooncake_store_connector.py	664	except Exception:	CODE
LOW	…/v1/kv_connector/unit/test_mooncake_store_connector.py	701	except Exception:	CODE
LOW	tests/v1/kv_connector/unit/test_rixl_gpu_mem_diag.py	36	except Exception:	CODE
LOW	…s/v1/kv_connector/nixl_integration/toy_proxy_server.py	233	except Exception as e:	CODE
MEDIUM	…s/v1/kv_connector/nixl_integration/toy_proxy_server.py	238	print(f"Error occurred in disagg prefill proxy server - {api} endpoint")	CODE
MEDIUM	…/kv_connector/nixl_integration/test_disagg_accuracy.py	151	print(f"Error writing to file: {e}")	CODE
MEDIUM	…/kv_connector/nixl_integration/test_disagg_accuracy.py	160	print(f"Error writing to file: {e}")	CODE
LOW	tests/v1/determinism/test_online_batch_invariance.py	42	except Exception as e: # pragma: no cover	CODE
LOW	tests/v1/distributed/test_external_lb_dp.py	88	except Exception as e:	CODE
LOW	tests/v1/distributed/test_external_lb_dp.py	115	except Exception as e:	CODE
MEDIUM	tests/v1/distributed/test_external_lb_dp.py	116	print(f"Error stopping servers: {e}")	CODE
LOW	tests/v1/distributed/test_internal_lb_dp.py	184	except Exception as e:	CODE
LOW	tests/v1/distributed/test_internal_lb_dp.py	212	except Exception as e:	CODE
MEDIUM	tests/v1/distributed/test_internal_lb_dp.py	213	print(f"Error stopping servers: {e}")	CODE
569 more matches not shown…

Modern Structural Boilerplate389 hits · 394 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	33	logger = logging.getLogger(__name__)	CODE
LOW	tests/v1/kv_offload/tiering/p2p/p2p_connector_proxy.py	32	logger = logging.getLogger(__name__)	CODE
LOW⚡	tests/v1/kv_connector/unit/test_multi_connector.py	116	def update_state_after_alloc(self, request, blocks, num_tokens) -> None:	CODE
LOW	tests/v1/kv_connector/unit/test_multi_connector.py	86	def update_state_after_alloc(self, request, blocks, num_tokens) -> None:	CODE
LOW⚡	…/v1/kv_connector/unit/test_handshake_pp_aggregation.py	136	def update_state_after_alloc(self, request: Any, blocks: Any, num_external_tokens: int) -> None:	CODE
LOW⚡	…/v1/kv_connector/unit/test_handshake_pp_aggregation.py	142	def set_xfer_handshake_metadata(self, metadata: dict[int, KVConnectorHandshakeMetadata]) -> None:	CODE
LOW	…/v1/kv_connector/unit/offloading_connector/conftest.py	7	__all__ = ["request_runner"]	CODE
LOW	…s/v1/kv_connector/nixl_integration/toy_proxy_server.py	15	logger = logging.getLogger(__name__)	CODE
LOW	tests/v1/worker/test_gpu_worker_weight_transfer.py	29	def update_weights(self, update_info: dict) -> None:	CODE
LOW	tests/v1/engine/test_core_engine_actor_manager.py	43	def _set_visible_devices(self, aphrodite_config: Any, local_dp_rank: int) -> None:	CODE
LOW	tests/kernels/moe/test_marlin_vs_trtllm_mxint4.py	71	__all__ = [	CODE
LOW⚡	tests/distributed/eplb_utils.py	73	def set_env_vars_and_device(env: dict[str, str]) -> None:	CODE
LOW	…/aphrodite_test_utils/aphrodite_test_utils/__init__.py	11	__all__ = ["blame", "BlameResult", "monitor", "MonitoredValues"]	CODE
LOW	tests/models/language/generation/test_hybrid.py	68	def _set_conv_state_layout(monkeypatch, layout: str) -> None:	CODE
LOW	…/models/multimodal/generation/vlm_utils/model_utils.py	37	logger = logging.getLogger(__name__)	CODE
LOW	tests/parser/engine/replay_harness.py	82	def set_vocab(self, vocab: dict[str, int]) -> None:	CODE
LOW	tests/entrypoints/openai/test_dp_supervisor.py	487	async def _set_healthy(port: int, use_ssl: bool = False) -> None:	CODE
LOW	tests/entrypoints/openai/test_dp_supervisor.py	497	async def _set_unhealthy(port: int, use_ssl: bool = False) -> None:	CODE
LOW	tests/entrypoints/openai/responses/conftest.py	15	logger = logging.getLogger(__name__)	CODE
LOW	…/entrypoints/openai/responses/test_parsable_context.py	21	logger = logging.getLogger(__name__)	CODE
LOW	tests/entrypoints/openai/responses/test_harmony.py	30	logger = logging.getLogger(__name__)	CODE
LOW	benchmarks/disagg_benchmarks/nixl_proxy_server.py	15	logger = logging.getLogger(__name__)	CODE
LOW	…marks/disagg_benchmarks/disagg_prefill_proxy_server.py	15	logger = logging.getLogger(__name__)	CODE
LOW	benchmarks/multi_turn/bench_utils.py	28	logger = logging.getLogger(__name__)	CODE
LOW	aphrodite/envs.py	521	logger = logging.getLogger(__name__)	CODE
LOW	aphrodite/env_override.py	393	def _update_scheduler_patched(self) -> None:	CODE
LOW	aphrodite/sampling_params.py	796	def update_from_tokenizer(self, tokenizer: TokenizerLike) -> None:	CODE
LOW	aphrodite/__init__.py	76	__all__ = [	CODE
LOW	aphrodite/v1/request.py	237	def update_block_hashes(self) -> None:	CODE
LOW	aphrodite/v1/kv_offload/tiering/example/manager.py	30	logger = logging.getLogger(__name__)	CODE
LOW	aphrodite/v1/kv_offload/tiering/p2p/data/__init__.py	6	__all__ = [	CODE
LOW	aphrodite/v1/kv_offload/tiering/p2p/control/__init__.py	12	__all__ = [	CODE
LOW	aphrodite/v1/kv_offload/tiering/p2p/session/__init__.py	10	__all__ = [	CODE
LOW	aphrodite/v1/kv_offload/tiering/fs/io.py	9	logger = logging.getLogger(__name__)	CODE
LOW	aphrodite/v1/metrics/stats.py	301	def update_from_output(self, prefill_stats: PrefillStats) -> None:	CODE
LOW	aphrodite/v1/attention/backends/flash_attn_diffkv.py	41	def set_head_size_v(cls, head_size_v: int) -> None:	CODE
LOW	aphrodite/v1/attention/backends/triton_attn_diffkv.py	81	def set_head_size_v(cls, head_size_v: int) -> None:	CODE
LOW	aphrodite/v1/attention/backends/mla/prefill/__init__.py	10	__all__ = [	CODE
LOW	aphrodite/v1/attention/ops/triton_decode_attention.py	42	logger = logging.getLogger(__name__)	CODE
LOW	aphrodite/v1/core/sched/async_scheduler.py	19	def _update_after_schedule(self, scheduler_output: SchedulerOutput) -> None:	CODE
LOW	aphrodite/v1/core/sched/interface.py	109	def update_draft_token_ids(self, draft_token_ids: "DraftTokenIds") -> None:	CODE
LOW	aphrodite/v1/core/sched/interface.py	202	def set_pause_state(self, pause_state: PauseState) -> None:	CODE
LOW⚡	aphrodite/v1/core/sched/scheduler.py	2256	def _update_waiting_for_remote_kv(self, request: Request) -> None:	CODE
LOW	aphrodite/v1/core/sched/scheduler.py	1094	def _update_after_schedule(self, scheduler_output: SchedulerOutput) -> None:	CODE
LOW	aphrodite/v1/core/sched/scheduler.py	1141	def _update_request_as_session(self, session: Request, update: StreamingUpdate) -> None:	CODE
LOW	aphrodite/v1/core/sched/scheduler.py	1802	def update_draft_token_ids(self, draft_token_ids: DraftTokenIds) -> None:	CODE
LOW	aphrodite/v1/core/sched/scheduler.py	1988	def set_pause_state(self, pause_state: PauseState) -> None:	CODE
LOW	aphrodite/v1/spec_decode/extract_hidden_states.py	80	def set_eplb_state(self, eplb_state: EplbState) -> None:	CODE
LOW⚡	aphrodite/v1/spec_decode/gemma4.py	61	def set_per_group_block_table(self, gid: int, block_table: torch.Tensor) -> None:	CODE
LOW	aphrodite/v1/spec_decode/gemma4.py	115	def _setup_centroids_cuda_graphs(self) -> None:	CODE
LOW	aphrodite/v1/spec_decode/llm_base_proposer.py	312	def set_eplb_state(self, eplb_state: EplbState) -> None:	CODE
LOW	aphrodite/v1/sample/thinking_budget_state.py	224	def _update_think_state(self, state: dict[str, Any]) -> None:	CODE
LOW	aphrodite/v1/sample/logits_processor/__init__.py	317	__all__ = [	CODE
LOW	aphrodite/v1/executor/aphrodite_net_devices.py	135	def set_worker_gpu_nic_mapping(local_rank: int) -> None:	CODE
LOW	aphrodite/v1/executor/aphrodite_net_devices.py	213	def set_worker_net_device(local_rank: int, aphrodite_config: AphroditeConfig) -> None:	CODE
LOW	aphrodite/v1/executor/__init__.py	6	__all__ = ["Executor", "UniProcExecutor"]	CODE
LOW	aphrodite/v1/worker/gpu_worker.py	417	def update_config(self, overrides: dict[str, Any]) -> None:	CODE
LOW	aphrodite/v1/worker/gpu_worker.py	661	def update_max_model_len(self, max_model_len: int) -> None:	CODE
LOW	aphrodite/v1/worker/gpu_worker.py	880	def _set_draft_weight_update_target(self) -> None:	CODE
LOW	aphrodite/v1/worker/gpu_worker.py	1159	def update_weights(self, update_info: dict) -> None:	CODE
329 more matches not shown…

Redundant / Tautological Comments223 hits · 340 pts

Severity	File	Line	Snippet	Context
LOW	tools/pre_commit/generate_attention_backend_docs.py	524	# Check if this is an MLA backend by parent class or naming	COMMENT
LOW	tools/pre_commit/generate_attention_backend_docs.py	1024	# Check if this is the "if use_mla:" branch	COMMENT
LOW	tests/conftest.py	387	# Set this to avoid hanging issue	COMMENT
LOW	tests/conftest.py	898	# Set this to avoid hanging issue	COMMENT
LOW	tests/utils.py	506	os.kill(spid, 0) # Check if still alive	CODE
LOW	tests/test_config.py	589	# Check if LONGCHAT_ROPE_PARAMETERS entries are in longchat_model_config	COMMENT
LOW	tests/v1/attention/test_mla_backends.py	978	# Set num_speculative_tokens to query_len - 1	COMMENT
LOW	tests/v1/attention/test_sparse_mla_backends.py	565	# Set some to -1 to test masking	COMMENT
LOW	tests/v1/attention/test_sparse_mla_backends.py	569	# Set some to out of bounds	COMMENT
LOW	tests/v1/attention/test_sparse_mla_backends.py	621	# Set some to -1 to test masking	COMMENT
LOW	tests/v1/attention/test_sparse_mla_backends.py	625	# Set some to out of bounds	COMMENT
LOW	tests/v1/core/test_scheduler.py	2294	# Verify if position length is identical	COMMENT
LOW	tests/v1/core/test_scheduler.py	3176	# Check if scheduled_encoder_inputs is empty as expected	COMMENT
LOW	tests/v1/core/test_scheduler.py	3512	# Set up to test different encoder cache existence scenario after preemption	COMMENT
LOW	tests/v1/core/test_scheduler.py	3820	# Set up to test different encoder cache existence scenario after preemption	COMMENT
LOW	tests/v1/core/utils.py	247	# Verify if position length is identical	COMMENT
LOW	…extract_hidden_states_integration/predictable_llama.py	81	# Check if we need auxiliary hidden states	COMMENT
LOW	…nnector/nixl_integration/config_sweep_accuracy_test.sh	98	# Check if cross-layers is enabled (non-empty)	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	270	# Check if tokens match first	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	526	# Check if tokens match first	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	738	# Check if tokens match	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	756	# Check if logprobs match bitwise	COMMENT
LOW	tests/v1/spec_decode/test_acceptance_length.py	108	# Check if get_valid_backends is actually defined in the platform class	COMMENT
LOW	…1/ec_connector/integration/run_epd_correctness_test.sh	25	# Set 1 to use multimodal prompts; else to use text-only	COMMENT
LOW	…ts/v1/ec_connector/integration/test_epd_correctness.py	211	# Check if server is ready	COMMENT
LOW	tests/v1/e2e/spec_decode/test_async_spec_decode.py	34	# Increment counter	COMMENT
LOW	tests/v1/engine/test_engine_core_client.py	303	# Check if all request IDs in outputs have finished	COMMENT
LOW	tests/v1/engine/utils.py	131	# Check if the sampled_token_id occurs in choice_tensor[1:]	COMMENT
LOW	tests/utils_/test_network_utils.py	79	# Check if IPv6 is supported by trying to create an IPv6 socket	COMMENT
LOW⚡	tests/tool_parsers/test_mistral_tool_parser.py	127	# Check if the slice from the current position matches the target sequence	COMMENT
LOW	tests/kernels/moe/test_moe_layer.py	1751	# Check if enough GPUs available	COMMENT
LOW⚡	tests/kernels/moe/test_rocm_aiter_topk.py	26	# Check if aiter package is installed	COMMENT
LOW⚡	tests/kernels/moe/test_rocm_aiter_topk.py	35	# Check if the op exists in torch.ops.aphrodite	COMMENT
LOW⚡	tests/kernels/moe/test_rocm_aiter_topk.py	38	# Check if the op is callable	COMMENT
LOW⚡	tests/kernels/moe/test_rocm_aiter_topk.py	44	# Check if the op exists in torch.ops.aphrodite	COMMENT
LOW⚡	tests/kernels/moe/test_rocm_aiter_topk.py	47	# Check if the op is callable	COMMENT
LOW	tests/evals/gsm8k/gsm8k_eval.py	336	# Print results to terminal	COMMENT
LOW	tests/distributed/test_eplb_execute.py	146	# Check if the weights are correct	COMMENT
LOW	tests/model_executor/test_qwen3_omni.py	29	# Check if it's a special token that should be compressed	COMMENT
LOW	…model_loader/runai_streamer_loader/test_runai_utils.py	56	# Read the file in chunks to handle large files efficiently	COMMENT
LOW⚡	tests/models/multimodal/generation/test_maverick.py	59	# Print the outputs	COMMENT
LOW	tests/quantization/test_gptq_v2.py	43	# Check if gptq_v2 format is correctly loaded	COMMENT
LOW	tests/quantization/test_gptq_v2.py	105	# Print the output sequences if failed	COMMENT
LOW	tests/compile/test_config.py	50	# Check if get_raw_stream exists in builtins	COMMENT
LOW	tests/compile/fusions_e2e/conftest.py	42	# Print the outputs.	COMMENT
LOW	tests/compile/fullgraph/test_full_graph.py	227	# Print the outputs.	COMMENT
LOW	tests/entrypoints/llm/offline_mode/test_offline_mode.py	71	# Set HF to offline mode and ensure we can still construct an LLM	COMMENT
LOW	tests/entrypoints/llm/offline_mode/test_offline_mode.py	138	# Set HF to offline mode and ensure we can still construct an LLM	COMMENT
LOW	tests/entrypoints/serve/utils/test_request_logger.py	95	# Set max_log_len to 10	COMMENT
LOW	…rve/sagemaker/test_sagemaker_middleware_integration.py	314	# Check if environment variable middleware was applied	COMMENT
LOW	tests/entrypoints/openai/completion/test_shutdown.py	100	# Check if any choice has finish_reason='abort'	COMMENT
LOW	tests/rocm/aiter/test_grouped_quant.py	21	# Check if aiter package is installed	COMMENT
LOW	tests/rocm/aiter/test_mla_fp8_support_check.py	107	# Set cache to True	COMMENT
LOW	tests/lora/test_transformers_model.py	39	# Print the outputs.	COMMENT
LOW	tests/lora/test_gptoss_tp.py	79	# Print the outputs.	COMMENT
LOW	tests/lora/test_qwen3moe_tp.py	55	# Print the outputs.	COMMENT
LOW	tests/lora/test_chatglm3_tp.py	38	# Print the outputs.	COMMENT
LOW	tests/lora/test_deepseekv2_tp.py	28	# Print the outputs.	COMMENT
LOW	tests/lora/test_minicpmv_tp.py	51	# Print the outputs.	COMMENT
LOW	tests/lora/test_mixtral.py	21	# Print the outputs.	COMMENT
163 more matches not shown…

AI Structural Patterns330 hits · 308 pts

Severity	File	Line	Context
LOW	tests/conftest.py	373	CODE
LOW	tests/conftest.py	411	CODE
LOW	tests/conftest.py	882	CODE
LOW	tests/v1/kv_offload/tiering/p2p/test_manager.py	382	CODE
LOW	tests/v1/attention/utils.py	169	CODE
LOW	tests/v1/attention/test_attention_backends.py	288	CODE
LOW	tests/v1/core/test_kv_cache_utils.py	112	CODE
LOW	tests/v1/core/test_scheduler.py	2113	CODE
LOW	tests/v1/core/test_scheduler.py	2234	CODE
LOW	tests/v1/core/utils.py	45	CODE
LOW	tests/v1/core/utils.py	196	CODE
LOW	tests/v1/kv_connector/unit/test_moriio_connector.py	225	CODE
LOW	tests/v1/kv_connector/unit/test_nixl_push_connector.py	339	CODE
LOW	tests/v1/kv_connector/unit/utils.py	77	CODE
LOW	tests/v1/kv_connector/unit/utils.py	177	CODE
LOW	tests/v1/sample/test_rejection_sampler.py	65	CODE
LOW	tests/v1/e2e/general/test_mamba_prefix_cache.py	161	CODE
LOW	tests/kernels/utils.py	274	CODE
LOW	tests/kernels/utils.py	817	CODE
LOW	tests/kernels/attention/test_mha_attn.py	117	CODE
LOW	tests/kernels/attention/test_attention.py	52	CODE
LOW	…sts/kernels/attention/test_triton_prefill_attention.py	71	CODE
LOW	tests/kernels/mamba/test_mamba_ssm.py	112	CODE
LOW	tests/kernels/quantization/test_nvfp4_quant.py	81	CODE
LOW	tests/kernels/moe/test_deepep_v2_moe.py	187	CODE
LOW	tests/kernels/moe/test_moe_permute_unpermute.py	104	CODE
LOW	tests/kernels/moe/test_moe.py	1499	CODE
LOW	tests/kernels/moe/utils.py	50	CODE
LOW	tests/kernels/moe/test_moe_layer.py	928	CODE
LOW	tests/kernels/moe/test_moe_layer.py	1027	CODE
LOW	tests/evals/mrcr/mrcr_eval.py	159	CODE
LOW	tests/evals/gsm8k/gsm8k_eval.py	202	CODE
LOW	tests/distributed/test_multiproc_executor.py	23	CODE
LOW	…/models/multimodal/generation/vlm_utils/model_utils.py	552	CODE
LOW	tests/models/transformers/fusers/test_moe.py	148	CODE
LOW	tests/quantization/reference_mxfp4.py	76	CODE
LOW	tests/quantization/reference_mxfp4.py	211	CODE
LOW	tests/compile/test_graph_partition.py	96	CODE
LOW	tests/compile/fullgraph/test_toy_llama.py	151	CODE
LOW	tests/compile/passes/test_silu_mul_quant_fusion.py	128	CODE
LOW	tests/compile/passes/test_fusion.py	466	CODE
LOW	tests/benchmarks/test_random_dataset.py	171	CODE
LOW	tests/entrypoints/openai/utils.py	135	CODE
LOW	examples/pooling/token_embed/jina_reranker_v3_online.py	22	CODE
LOW	benchmarks/kernels/benchmark_mrope.py	78	CODE
LOW	aphrodite/sampling_params.py	436	CODE
LOW	aphrodite/_custom_ops.py	3005	CODE
LOW	aphrodite/_custom_ops.py	3711	CODE
LOW	aphrodite/_xpu_ops.py	722	CODE
LOW	aphrodite/_xpu_ops.py	797	CODE
LOW	aphrodite/forward_context.py	198	CODE
LOW	aphrodite/forward_context.py	246	CODE
LOW	aphrodite/_aiter_ops.py	106	CODE
LOW	aphrodite/_aiter_ops.py	166	CODE
LOW	aphrodite/_aiter_ops.py	195	CODE
LOW	aphrodite/_aiter_ops.py	232	CODE
LOW	aphrodite/_aiter_ops.py	462	CODE
LOW	aphrodite/_aiter_ops.py	520	CODE
LOW	aphrodite/_aiter_ops.py	2154	CODE
LOW	aphrodite/_aiter_ops.py	2204	CODE
270 more matches not shown…

Structural Annotation Overuse158 hits · 291 pts

Severity	File	Line	Snippet	Context
LOW⚡	build_and_upload_docker.sh	23	# Step 1: Build the wheel	COMMENT
LOW⚡	build_and_upload_docker.sh	39	# Step 2: Build the final Docker image using the wheel	COMMENT
LOW⚡	build_and_upload_docker.sh	47	# Step 3: Tag and push the Docker image	COMMENT
LOW⚡	tests/v1/kv_offload/tiering/test_async_lookup.py	134	# Step 1: lookup key 1, flush	COMMENT
LOW⚡	tests/v1/kv_offload/tiering/test_async_lookup.py	140	# Step 2: lookup keys 2 and 3, flush	COMMENT
LOW	tests/v1/logits_processors/test_correctness.py	1114	# Step 1: think-start token appears.	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5010	# Step 1: A's load is admitted; B's is held back by the reservation (B never	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5020	# Step 2: nothing changes until A's recv lands.	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5029	# Step 3: A makes forward progress straight to RUNNING - no preemption was	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	551	# Step 1: Get events from lmcache engine	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	558	# Step 2: Update connector output (simulate receiving from worker)	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	564	# Step 3: Take events	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	401	# Step 2: 5 blocks are in use (2 new for remote blocks).	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	409	# Step 3: finish recving (5 blocks in use)	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	418	# Step 4: try to schedule, remote request is put to running list	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	426	# Step 5: Remote request will be put back to waiting list	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	434	# Step 6: finish the request, free it.	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	441	# Step 7: now we can schedule (with 2 blocks computed),	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	511	# Step 3: finish the request, free it.	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	518	# Step 4: now we can initiate KV transfer (with 2 blocks computed).	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	526	# Step 5: finish recving (5 blocks in use)	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	533	# Step 6: schedule remote request	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	540	# Step 7: free everything.	COMMENT
LOW	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	457	# Step 8: free everything.	COMMENT
LOW	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	500	# Step 2: 3 blocks are in use,	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	665	# Step 1: Run decode and collect logprobs	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	681	# Step 2: For each token position, run prefill and compare	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	369	# Step 2: Schedule creates NewRequestData	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	379	# Step 3: Simulate model runner caching the prompt_token_ids	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	432	# Step 7: Schedule again - now request uses cached state	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	440	# Step 8: Calculate num_tokens like gpu_model_runner.py:1284 does	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	491	# Step 12: Add new streaming request with seq_id=1	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	408	# Step 6: Verify request state after Cycle 1	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	472	# Step 11: Verify request transitioned to WAITING_FOR_STREAMING_REQ	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	503	# Step 13: Scheduler schedules the updated session	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	516	# Step 14: Model runner caches NEW prompt_token_ids reference	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	527	# Step 15: FINAL CRITICAL VERIFICATION	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	84	# Step 1: Add initial request with 3 prompt tokens, all computed	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	100	# Step 2: Create streaming update with extended prompt	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	116	# Step 3: Verify no free_indices leak (old slot recycled)	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	153	# Step 1: Add initial request with one audio feature	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	173	# Step 2: Create streaming update with additional multimodal feature	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	190	# Step 3: Verify no free_indices leak	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	59	# Step 1: Create initial request state with some computed tokens	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	77	# Step 2: Create new request data with extended prompt	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	96	# Step 3: Update the request	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	99	# Step 4: Verify the request state was updated correctly	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	129	# Step 1: Create initial request state with one multimodal feature	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	154	# Step 2: Create new request data with additional multimodal feature	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	174	# Step 3: Update the request	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	177	# Step 4: Verify the request state was updated correctly	COMMENT
LOW⚡	tests/v1/ec_connector/integration/README.md	61	### Step 1: Baseline	COMMENT
LOW⚡	tests/v1/ec_connector/integration/README.md	68	### Step 2: EPD (1E + 1PD)	COMMENT
LOW⚡	tests/v1/ec_connector/integration/README.md	76	### Step 3: EPD (1E + 1P + 1D)	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	459	# Step 1: Run baseline	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	462	# Step 2: Test 1E + 1PD	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	465	# Step 3: Test baseline 1P + 1D	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	468	# Step 4: Test 1E + 1P + 1D	COMMENT
LOW	tests/kernels/test_fused_inv_rope_fp8_quant.py	630	# Step 1: In-place CUDA RoPE (same as production)	COMMENT
LOW	tests/kernels/test_fused_inv_rope_fp8_quant.py	642	# Step 2: Reshape + quant + reshape (same as production)	COMMENT
98 more matches not shown…

Cross-Language Confusion54 hits · 285 pts

Severity	File	Line	Snippet	Context
HIGH	csrc/cpu/generate_cpu_attn_dispatch.py	154	(__riscv_v_min_vlen == 128 \|\| __riscv_v_min_vlen == 256)	CODE
HIGH	csrc/cpu/generate_cpu_attn_dispatch.py	223	"&& (__riscv_v_min_vlen == 128 \|\| __riscv_v_min_vlen == 256)",	CODE
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	490	"M > 256 && K <= 16384 && N <= 4096": ((128, 128), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	493	"M > 128 && K <= 4096 && N <= 4096": ((128, 64), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	494	"M > 128 && K <= 8192 && N <= 8192": ((128, 128), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	497	"M > 64 && K <= 4069 && N <= 4069": ((128, 32), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	498	"M > 64 && K <= 4069 && N <= 8192": ((128, 64), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	499	"M > 64 && K >= 8192 && N >= 12288": ((256, 128), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	502	"M > 32 && K <= 6144 && N <= 6144": ((128, 16), (1, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	503	"M > 32 && K >= 16384 && N >= 12288": ((256, 64), (2, 1, 1)),	STRING
HIGH⚡	csrc/libtorch_stable/quantization/machete/generate.py	506	"M > 16 && K <= 12288 && N <= 8192": ((128, 32), (2, 1, 1)),	STRING
HIGH	tests/v1/attention/test_trtllm_attention_integration.py	142	# Randomly permute blocks (starting from block 1; block 0 is null).	COMMENT
HIGH	tests/v1/attention/test_mla_backends.py	237	# Permute the context blocks (excluding block 0 which is null)	COMMENT
HIGH	tests/v1/attention/test_attention_backends.py	149	# Permute the context blocks (excluding block 0 which is null)	COMMENT
HIGH⚡	tests/v1/core/test_scheduler.py	2918	num_blocks=5, # Can hold 64 tokens (first block is null)	CODE
HIGH⚡	tests/v1/core/test_scheduler.py	3675	num_blocks=15, # can hold 244 tokens with 14 blocks (first block is null)	CODE
HIGH	tests/v1/core/test_scheduler.py	3902	num_blocks=11, # Can hold 160 tokens (first block is null)	CODE
HIGH	tests/v1/cudagraph/test_cudagraph_mode.py	65	# when above code raises, `llm` may be undefined, so we need to catch that	COMMENT
HIGH	tests/v1/cudagraph/test_cudagraph_mode.py	115	# when above code raises, `llm` may be undefined, so we need to catch that	COMMENT
HIGH	…_connector/unit/offloading_connector/test_scheduler.py	2393	# 4 GPU blocks: block 0 is null, blocks 1-3 are usable.	COMMENT
HIGH	tests/tool_parsers/test_granite_tool_parser.py	41	"null_field": null,	CODE
HIGH	tests/tool_parsers/test_granite_20b_fc_tool_parser.py	33	"null_field": null,	CODE
HIGH⚡	tests/tool_parsers/test_olmo3_tool_parser.py	34	"role=null, "	CODE
HIGH⚡	tests/tool_parsers/test_olmo3_tool_parser.py	43	'"role": null, '	CODE
HIGH⚡	tests/tool_parsers/test_pythonic_tool_parser.py	35	'"role": null, '	CODE
HIGH	tests/tool_parsers/test_deepseekv3_tool_parser.py	41	""""bool_field": true, "null_field": null, """	STRING
HIGH	tests/tool_parsers/test_internlm2_tool_parser.py	52	"null_field": null,	CODE
HIGH	tests/tool_parsers/test_llama4_pythonic_tool_parser.py	35	'"role": null, '	CODE
HIGH	tests/tool_parsers/test_phi4mini_tool_parser.py	49	"null_field": null,	CODE
HIGH⚡	tests/tool_parsers/test_lfm2_tool_parser.py	38	'"role": null, '	CODE
HIGH	tests/tool_parsers/test_longcat_tool_parser.py	53	"null_field": null,	CODE
HIGH⚡	tests/tool_parsers/test_hunyuan_a13b_tool_parser.py	41	'<tool_calls>[{"name": "get_weather", "arguments": {"city": "San Francisco", "metric": "celsius"}}, {"name":	CODE
HIGH	tests/tool_parsers/test_qwen3coder_tool_parser.py	570	# Multi non-null: anyOf[string, integer, null] → first non-null is string	COMMENT
HIGH⚡	tests/parser/engine/test_qwen3.py	257	' "multiSelect": false, "answer": null}]'	CODE
HIGH⚡	tests/parser/engine/test_qwen3.py	267	"questions": '[{"question": "Pick a color", "multiSelect": false, "answer": null}]',	CODE
HIGH⚡	tests/parser/engine/test_qwen3.py	1034	' "multiSelect": false, "answer": null}]'	CODE
HIGH	tests/parser/engine/test_qwen3.py	1053	' "multiSelect": false, "answer": null}]',	CODE
HIGH⚡	tests/parser/engine/test_parser_engine.py	598	result = engine._fix_arg_types('{"val": null}', "f")	CODE
HIGH⚡	tests/parser/engine/test_gemma4_streaming_reasoning.py	1138	text = "<\|tool_call>call:configure{value:null}<tool_call\|>"	CODE
HIGH⚡	tests/parser/engine/test_gemma4_streaming_reasoning.py	1144	text = "<\|tool_call>call:configure{label:null}<tool_call\|>"	CODE
HIGH	…e_out/token_in_token_out/test_return_routed_experts.py	36	'{"sliding_window": null}',	CODE
HIGH	tests/entrypoints/openai/test_return_routed_experts.py	33	'{"sliding_window": null}',	CODE
HIGH	aphrodite/_custom_ops.py	2101	assert k_times_2 % 2 == 0, "input width must be even (gate \|\| up layout)"	CODE
HIGH	aphrodite/_custom_ops.py	2206	assert k_times_2 % 2 == 0, "input width must be even (gate \|\| up layout)"	CODE
HIGH	aphrodite/_custom_ops.py	2083	input_tensor: The input tensor with gate \|\| up layout [m_topk, k*2]	STRING
HIGH	aphrodite/v1/core/single_type_kv_cache_manager.py	1051	# result [null] [null] ... [null] [hit block 1 (1st block contain	COMMENT
HIGH	aphrodite/v1/core/single_type_kv_cache_manager.py	418	every (non-null) block — the default for full attention.	STRING
HIGH	aphrodite/v1/core/single_type_kv_cache_manager.py	1004	[null, null, block 3], otherwise, we return [null, null]	STRING
HIGH	aphrodite/v1/core/single_type_kv_cache_manager.py	1009	we return 4 blocks[null, null, null, null]	STRING
HIGH	aphrodite/tool_parsers/utils.py	432	(null, true, false) that some models produce instead of Python	STRING
HIGH	aphrodite/metal/mlx_backend/mla_cache.py	11	Each token's cache entry is a combined latent vector [kv_norm \|\| k_pe]:	STRING
HIGH	aphrodite/metal/paged_attention_backend/mla.py	29	latent = [kv_norm \|\| k_pe_roped] # kv_lora_rank + qk_rope_head_dim dims	STRING
HIGH	…ibuted/kv_transfer/kv_connector/v1/flexkv_connector.py	47	cd FlexKV && bash build.sh	STRING
HIGH	aphrodite/benchmarks/datasets/datasets.py	3953	# undefined), language (en), transcription directive (en), punctuation	STRING

Verbosity Indicators137 hits · 257 pts

Severity	File	Line	Snippet	Context
LOW⚡	build_and_upload_docker.sh	23	# Step 1: Build the wheel	COMMENT
LOW⚡	build_and_upload_docker.sh	39	# Step 2: Build the final Docker image using the wheel	COMMENT
LOW⚡	build_and_upload_docker.sh	47	# Step 3: Tag and push the Docker image	COMMENT
LOW⚡	tests/v1/kv_offload/tiering/test_async_lookup.py	134	# Step 1: lookup key 1, flush	COMMENT
LOW⚡	tests/v1/kv_offload/tiering/test_async_lookup.py	140	# Step 2: lookup keys 2 and 3, flush	COMMENT
LOW	tests/v1/logits_processors/test_correctness.py	1114	# Step 1: think-start token appears.	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5010	# Step 1: A's load is admitted; B's is held back by the reservation (B never	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5020	# Step 2: nothing changes until A's recv lands.	COMMENT
LOW⚡	tests/v1/core/test_scheduler.py	5029	# Step 3: A makes forward progress straight to RUNNING - no preemption was	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	551	# Step 1: Get events from lmcache engine	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	558	# Step 2: Update connector output (simulate receiving from worker)	COMMENT
LOW⚡	tests/v1/kv_connector/unit/test_lmcache_connector.py	564	# Step 3: Take events	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	401	# Step 2: 5 blocks are in use (2 new for remote blocks).	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	409	# Step 3: finish recving (5 blocks in use)	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	418	# Step 4: try to schedule, remote request is put to running list	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	426	# Step 5: Remote request will be put back to waiting list	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	434	# Step 6: finish the request, free it.	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	441	# Step 7: now we can schedule (with 2 blocks computed),	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	511	# Step 3: finish the request, free it.	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	518	# Step 4: now we can initiate KV transfer (with 2 blocks computed).	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	526	# Step 5: finish recving (5 blocks in use)	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	533	# Step 6: schedule remote request	COMMENT
LOW⚡	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	540	# Step 7: free everything.	COMMENT
LOW	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	457	# Step 8: free everything.	COMMENT
LOW	…/v1/kv_connector/unit/test_remote_prefill_lifecycle.py	500	# Step 2: 3 blocks are in use,	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	665	# Step 1: Run decode and collect logprobs	COMMENT
LOW	tests/v1/determinism/test_batch_invariance.py	681	# Step 2: For each token position, run prefill and compare	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	369	# Step 2: Schedule creates NewRequestData	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	379	# Step 3: Simulate model runner caching the prompt_token_ids	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	432	# Step 7: Schedule again - now request uses cached state	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	440	# Step 8: Calculate num_tokens like gpu_model_runner.py:1284 does	COMMENT
LOW⚡	tests/v1/streaming_input/test_scheduler_streaming.py	491	# Step 12: Add new streaming request with seq_id=1	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	408	# Step 6: Verify request state after Cycle 1	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	472	# Step 11: Verify request transitioned to WAITING_FOR_STREAMING_REQ	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	503	# Step 13: Scheduler schedules the updated session	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	516	# Step 14: Model runner caches NEW prompt_token_ids reference	COMMENT
LOW	tests/v1/streaming_input/test_scheduler_streaming.py	527	# Step 15: FINAL CRITICAL VERIFICATION	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	84	# Step 1: Add initial request with 3 prompt tokens, all computed	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	100	# Step 2: Create streaming update with extended prompt	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	116	# Step 3: Verify no free_indices leak (old slot recycled)	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	153	# Step 1: Add initial request with one audio feature	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	173	# Step 2: Create streaming update with additional multimodal feature	COMMENT
LOW	…/streaming_input/test_gpu_model_runner_v2_streaming.py	190	# Step 3: Verify no free_indices leak	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	59	# Step 1: Create initial request state with some computed tokens	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	77	# Step 2: Create new request data with extended prompt	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	96	# Step 3: Update the request	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	99	# Step 4: Verify the request state was updated correctly	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	129	# Step 1: Create initial request state with one multimodal feature	COMMENT
LOW	…/v1/streaming_input/test_gpu_model_runner_streaming.py	154	# Step 2: Create new request data with additional multimodal feature	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	174	# Step 3: Update the request	COMMENT
LOW⚡	…/v1/streaming_input/test_gpu_model_runner_streaming.py	177	# Step 4: Verify the request state was updated correctly	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	459	# Step 1: Run baseline	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	462	# Step 2: Test 1E + 1PD	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	465	# Step 3: Test baseline 1P + 1D	COMMENT
LOW⚡	…1/ec_connector/integration/run_epd_correctness_test.sh	468	# Step 4: Test 1E + 1P + 1D	COMMENT
LOW	tests/kernels/test_fused_inv_rope_fp8_quant.py	630	# Step 1: In-place CUDA RoPE (same as production)	COMMENT
LOW	tests/kernels/test_fused_inv_rope_fp8_quant.py	642	# Step 2: Reshape + quant + reshape (same as production)	COMMENT
LOW⚡	…nts/multimodal/llm/test_mm_cache_external_injection.py	92	# Step 1: Normal requests to populate the cache	COMMENT
LOW⚡	…nts/multimodal/llm/test_mm_cache_external_injection.py	100	# Step 2: Use a second image to get valid expanded tokens and	COMMENT
LOW⚡	…ntrypoints/weight_transfer/test_weight_transfer_llm.py	258	# Step 1: Initialize weight transfer engine	COMMENT
77 more matches not shown…

Docstring Block Structure37 hits · 185 pts

Severity	File	Line	Snippet	Context
HIGH	benchmarks/benchmark_long_document_qa_throughput.py	70	Repeat each prompt in the list for a specified number of times. The order of prompts in the output list depends	STRING
HIGH	aphrodite/v1/attention/backends/registry.py	204	Register or override a backend implementation. Args: backend: The AttentionBackendEnum member to register	STRING
HIGH	aphrodite/v1/attention/backends/mla/prefill/registry.py	90	Register or override an MLA prefill backend implementation. Args: backend: The MLAPrefillBackendEnum member	STRING
HIGH	aphrodite/v1/core/single_type_kv_cache_manager.py	991	For chunked local attention, we need to find the longest cache hit prefix of the blocks that is not lon	STRING
HIGH	aphrodite/v1/structured_output/utils.py	314	Check if grammar appears to use Lark syntax. Args: grammar_str: Input grammar string Returns:	STRING
HIGH	aphrodite/v1/structured_output/utils.py	346	Convert a Lark grammar string to EBNF format. EBNF reference: https://github.com/ggerganov/llama.cpp/blob/	STRING
HIGH	aphrodite/v1/worker/utils.py	275	Select a block size that is supported by all backends and is a factor of kv_manager_block_size. If kv_mana	STRING
HIGH	aphrodite/tool_parsers/apertus_tool_parser.py	134	Buffers incoming delta chunks to prevent fragmentation of multi-token special tags. If a chunk	STRING
HIGH	aphrodite/tool_parsers/apertus_tool_parser.py	178	Extracts tool calls from a completely generated model response (Non-Streaming). Args: mode	STRING
HIGH	aphrodite/tool_parsers/apertus_tool_parser.py	274	Handles streaming chunks Args: previous_text: The complete model text generated prior to t	STRING
HIGH	aphrodite/tool_parsers/apertus_tool_parser.py	467	Calculates the exact string difference to safely append new tool parameters. This ensures characters l	STRING
HIGH	aphrodite/metal/metal_kernel_backend/attention_sdpa.py	166	Project ``x`` into Q/K/V with norms, RoPE and Gemma4 variants. Handles three Gemma4-specific branches: - **YOC	STRING
HIGH	aphrodite/metal/metal_kernel_backend/attention_sdpa.py	308	Zero-pad Q/K/V on the last axis up to ``cache_head_dim``. Variable head_dim models (e.g. Gemma4 sliding=256, full=5	STRING
HIGH	aphrodite/metal/v1/model_adapter.py	231	Return per-layer ``(kv_heads, head_dim)`` lists for Gemma4, else None. Gemma4 26B/31B mix sliding attention (``	STRING
HIGH	aphrodite/metal/stt/audio.py	57	Load an audio file and return mono samples at sample_rate Hz. Tries ``librosa`` first (handles WAV natively witho	STRING
HIGH	aphrodite/metal/stt/audio.py	90	Load audio via ffmpeg subprocess. Args: file_path: Path to the audio file. sample_rate: Target samp	STRING
HIGH	…ed/kv_transfer/kv_connector/v1/lmcache_mp_connector.py	709	Get number of new tokens that can be loaded from the external KV cache beyond the num_computed_tokens.	STRING
HIGH	…rodite/distributed/kv_transfer/kv_connector/v1/base.py	453	Get number of new tokens that can be loaded from the external KV cache beyond the num_computed_tokens.	STRING
HIGH	…ed/kv_transfer/kv_connector/v1/moriio/moriio_engine.py	276	Get remote allocation info for a request. Args: transfer_id:TransferId The request ID Retu	STRING
HIGH	…nector/v1/lmcache_integration/multi_process_adapter.py	194	Submit a new lookup request to LMCache if there is no ongoing request. Supports both token-based and h	STRING
HIGH	…nector/v1/lmcache_integration/multi_process_adapter.py	544	Check and get the finished store and retrieve requests. Args: finished_req_ids_from_engine	STRING
HIGH	aphrodite/distributed/weight_transfer/factory.py	81	Create a weight transfer engine instance. Args: config: Weight transfer configuration containing th	STRING
HIGH	aphrodite/distributed/weight_transfer/base.py	117	Construct typed init info from dict with validation. Args: init_dict: Dictionary containin	STRING
HIGH	aphrodite/distributed/weight_transfer/base.py	135	Construct typed update info from dict with validation. Args: update_dict: Dictionary conta	STRING
HIGH	aphrodite/model_executor/kernels/linear/__init__.py	518	Choose a _KernelT that can implement the given config for the given compute capability. Attempts to choose the	STRING
HIGH	aphrodite/model_executor/kernels/linear/__init__.py	692	Choose an MPLinearKernel that can implement the given config for the given compute capability. Attempts to cho	STRING
HIGH	…/model_executor/layers/fused_moe/expert_map_manager.py	322	Map global expert ID to local expert ID. Args: global_id: Global expert ID (0 to global_nu	STRING
HIGH	aphrodite/model_executor/layers/fla/ops/chunk.py	147	Args: q (torch.Tensor): Queries of shape `[B, T, H, K]`. k (torch.Tensor):	STRING
HIGH	…odite/model_executor/layers/fla/ops/fused_recurrent.py	509	Args: q (torch.Tensor): queries of shape `[B, T, H, K]`. k (torch.Tensor):	STRING
HIGH	aphrodite/model_executor/models/keye_vl1_5.py	77	Return num_patches per video. Args: grid_thw: Tensor with shape [N, 3] containing temporal, height, wi	STRING
HIGH	aphrodite/model_executor/models/isaac.py	228	Apply pixel shuffle to a packed vision sequence without unpacking per image. Args: x (`torch.Tensor`):	STRING
HIGH	aphrodite/multimodal/audio.py	92	Normalize audio to the specified format. This function handles channel reduction for multi-channel audio, suppo	STRING
HIGH	aphrodite/benchmarks/lib/ready_checker.py	25	Wait for an endpoint to become available before starting benchmarks. Args: request_func: The async req	STRING
HIGH	aphrodite/entrypoints/chat_utils.py	1338	Parses a given multi-modal content part based on its type. Args: part: A dict containing the content p	STRING
HIGH	aphrodite/entrypoints/speech_to_text/base/utils.py	18	Read an uploaded file enforcing a size limit before full materialization. The function first checks the Conte	STRING
HIGH	aphrodite/transformers_utils/processors/isaac.py	174	Convert normalized images into flattened ViT-style patches. Args: image (`torch.Tensor`): Tenso	STRING
HIGH	aphrodite/lora/resolver.py	69	Get a registered resolver instance by name. Args: resolver_name: Name of the resolver to get.	STRING

AI Slop Vocabulary76 hits · 164 pts

Severity	File	Line	Snippet	Context
LOW	csrc/libtorch_stable/quantization/machete/generate.py	413	# For now, we can just use the first accumulator type seen since	STRING
MEDIUM	tests/v1/logits_processors/test_custom_offline.py	28	# Create a mixture of requests which do and don't utilize the dummy logitproc	COMMENT
LOW	tests/v1/cudagraph/test_breakable_cudagraph.py	235	# Outside capture: decorator should just call through.	COMMENT
LOW⚡	tests/tool_parsers/test_mistral_tool_parser.py	133	# Otherwise, just add the current token and move to the next one	COMMENT
MEDIUM⚡	tests/renderers/test_sparse_tensor_validation.py	58	# explicitly so this fixture is robust to process-wide invariant-check state	COMMENT
MEDIUM	tests/kernels/mamba/test_mamba_mixer2.py	106	# - utilize mock patching to disable TP when	COMMENT
MEDIUM	tests/distributed/test_pynccl.py	373	# Essentially this is an all-gather operation.	COMMENT
MEDIUM	tests/distributed/test_context_parallel.py	46	# .buildkite/lm-eval-harness/configs/DeepSeek-V2-Lite-Chat.yaml	COMMENT
MEDIUM	tests/distributed/test_context_parallel.py	48	# .buildkite/lm-eval-harness/configs/Qwen2.5-1.5B-Instruct.yaml	COMMENT
LOW	tests/model_executor/test_qwen3_omni.py	41	# Regular token, just add it	COMMENT
MEDIUM	tests/models/registry.py	921	# Keep the init/schema harness eager (these tests never run forward).	COMMENT
MEDIUM	…s/compile/passes/distributed/test_fusion_all_reduce.py	717	# the call site). The patterns in this PR are robust to both Triton and	COMMENT
LOW	tests/lora/test_gptoss_tp.py	91	# For now just use TRITON_UNFUSED kernel	COMMENT
LOW	examples/offline_inference/ray_distributed_inference.py	71	# For tensor_parallel_size == 1, we simply set num_gpus=1.	COMMENT
MEDIUM	benchmarks/kernels/benchmark_silu_mul_fp8_quant.py	678	# Create comprehensive 3-way comparison plots	COMMENT
LOW	aphrodite/env_override.py	312	# functions just return True.	COMMENT
LOW	aphrodite/v1/attention/backends/flash_attn.py	316	# but for now just set it to `UNIFORM_BATCH` to get use to drop down	COMMENT
LOW	aphrodite/v1/attention/backends/utils.py	278	# then we can simply use a cdiv for the rest.	COMMENT
MEDIUM	aphrodite/v1/attention/ops/triton_decode_attention.py	358	# explicitly facilitate overlapping load/compute	COMMENT
LOW	aphrodite/v1/core/kv_cache_coordinator.py	457	# Single group; useless but just set ``use_eagle`` for consistency regardless.	COMMENT
MEDIUM	aphrodite/v1/core/encoder_cache_manager.py	314	# utilize the cache and this class will fold into EncoderCacheManager, as	COMMENT
LOW	aphrodite/v1/spec_decode/llm_base_proposer.py	560	# KV cache in sync, so just return an empty tensor.	COMMENT
LOW	aphrodite/v1/spec_decode/llm_base_proposer.py	1683	# Therefore, we can just return the logits.	COMMENT
MEDIUM	aphrodite/v1/worker/gpu_worker.py	758	# CUDAGraph memory size and may not utilize all gpu memory.	COMMENT
MEDIUM	aphrodite/v1/worker/gpu_worker.py	1019	# Generate the trace name by combining prefix with comprehensive rank suffix	COMMENT
LOW	aphrodite/v1/worker/gpu/model_runner.py	1240	# For piecewise and eager mode, just call model().	COMMENT
LOW	aphrodite/v1/engine/parallel_sampling.py	116	# If streaming, just return the current output	COMMENT
MEDIUM	aphrodite/tool_parsers/llama_tool_parser.py	219	# re-set stuff pertaining to progress in the current tool	COMMENT
MEDIUM	aphrodite/tool_parsers/jamba_tool_parser.py	182	# re-set stuff pertaining to progress in the current tool	COMMENT
MEDIUM	aphrodite/tool_parsers/granite_tool_parser.py	163	# re-set stuff pertaining to progress in the current tool	COMMENT
LOW	aphrodite/tool_parsers/xlam_tool_parser.py	492	# If we encounter an error, just return the delta text as regular content	COMMENT
MEDIUM	aphrodite/tool_parsers/granite_20b_fc_tool_parser.py	177	# re-set stuff pertaining to progress in the current tool	COMMENT
MEDIUM	aphrodite/tool_parsers/step3_tool_parser.py	51	# Explicit state flags for robust streaming	COMMENT
LOW	aphrodite/tokenizers/mistral.py	504	# if underlying tokenizer is sentencepiece, we just add "�".	COMMENT
LOW	aphrodite/platforms/cuda.py	727	# users can just use IR op priority directly	COMMENT
MEDIUM	aphrodite/distributed/utils.py	321	"""A robust barrier to synchronize all ranks.	STRING
MEDIUM	…ransfer/kv_connector/v1/mooncake/mooncake_connector.py	859	# Tasks can await async events, so a surplus (2x is a robust heuristic)	COMMENT
MEDIUM	aphrodite/config/model.py	308	"""Enable the custom cumem allocator to leverage advanced GPU memory	STRING
MEDIUM	aphrodite/config/parallel.py	574	# To make the initialization more robust we retry a few times	COMMENT
LOW	…odite/model_executor/layers/mamba/ops/causal_conv1d.py	179	# first chunk and does not have prior-token, so just set to 0	COMMENT
MEDIUM	…dite/model_executor/layers/fused_moe/modular_kernel.py	48	# The goal is to be able to utilize different communication mechanisms with	COMMENT
MEDIUM	…xecutor/layers/fused_moe/prepare_finalize/deepep_ll.py	38	# TODO (varun) : Optimize leverage num_tokens_per_expert counts	COMMENT
MEDIUM	…executor/layers/fused_moe/experts/fused_humming_moe.py	308	# Neighboring nodes are required to utilize distinct workspaces.	COMMENT
MEDIUM	aphrodite/model_executor/layers/quantization/fp8.py	280	# For GPUs that lack FP8 hardware support, we can leverage the Marlin	COMMENT
MEDIUM	…hrodite/model_executor/layers/quantization/modelopt.py	307	# Normalize quant_algo for robust matching (ModelOpt may emit lowercase).	COMMENT
MEDIUM	…odite/model_executor/layers/quantization/fbgemm_fp8.py	53	# For GPUs that lack FP8 hardware support, we can leverage the Marlin	COMMENT
MEDIUM	…executor/layers/quantization/utils/marlin_utils_fp4.py	107	# to fully utilize the E4M3 dynamic range (e.g., global_scale=1).	COMMENT
MEDIUM	…executor/layers/quantization/utils/marlin_utils_fp4.py	165	# For GPUs that lack FP4 hardware support, we can leverage the	COMMENT
MEDIUM	…executor/layers/quantization/utils/marlin_utils_fp8.py	60	# For GPUs that lack FP8 hardware support, we can leverage the	COMMENT
LOW	…odel_executor/layers/quantization/utils/quant_utils.py	390	# Unquantized layer: just return base weights	COMMENT
MEDIUM	…rs/quantization/compressed_tensors/transform/module.py	81	# do not fold into weight in order to utilize FWHT	COMMENT
MEDIUM	aphrodite/model_executor/models/deepseek_ocr.py	140	"""Example of overriding the wrapper class `__init__()` in order to utilize	STRING
MEDIUM	aphrodite/model_executor/models/deepseek_ocr.py	140	"""Example of overriding the wrapper class `__init__()` in order to utilize	STRING
LOW	aphrodite/model_executor/models/qwen3_asr.py	516	# No audio features, just return linear positions	COMMENT
LOW	…odite/model_executor/models/transformers/multimodal.py	190	# NOTE: we can't just set caching=False because base class method	COMMENT
LOW	aphrodite/models/deepseek_v4/xpu/xpu_sparse.py	51	# Profile run: no-op, just return q (no padding needed on XPU).	COMMENT
MEDIUM	…dels/deepseek_v4/nvidia/ops/fused_indexer_q_cutedsl.py	163	# all threads in a warp to be active since we utilize warp shuffle later.	COMMENT
LOW	aphrodite/reasoning/granite_reasoning_parser.py	182	# corrected; just return the delta text as normal content.	COMMENT
LOW	aphrodite/reasoning/hunyuan_a13b_reasoning_parser.py	88	# this id is not part of content, so just return [] here.	COMMENT
LOW	aphrodite/reasoning/olmo3_reasoning_parser.py	251	# this id is not part of content, so just return [] here.	COMMENT
16 more matches not shown…

Hallucination Indicators10 hits · 105 pts

Severity	File	Line	Snippet	Context
CRITICAL	rust/src/cmd/src/cli/unsupported.rs	138	/// - `aphrodite.entrypoints.cli.serve.ServeSubcommand.subparser_init(...)`	COMMENT
CRITICAL	tests/v1/e2e/general/test_mamba_prefix_cache.py	726	assert engine.llm_engine.engine_core.engine_core.scheduler.reset_prefix_cache()	CODE
CRITICAL	tests/v1/e2e/general/test_mamba_prefix_cache.py	938	assert engine.llm_engine.engine_core.engine_core.scheduler.reset_prefix_cache()	CODE
CRITICAL	tests/v1/e2e/general/test_mamba_prefix_cache.py	957	assert engine.llm_engine.engine_core.engine_core.scheduler.reset_prefix_cache()	CODE
CRITICAL	tests/distributed/test_torchrun_example_moe.py	80	params = list(llm.llm_engine.model_executor.driver_worker.worker.model_runner.model.parameters())	CODE
CRITICAL	tests/distributed/test_torchrun_example.py	71	params = list(llm.llm_engine.model_executor.driver_worker.worker.model_runner.model.parameters())	CODE
CRITICAL	tests/models/language/generation/test_gemma.py	19	lambda self: self.model_runner.model.language_model.model.normalizer.cpu().item() # noqa: E501	CODE
CRITICAL	tests/models/language/generation/test_gemma.py	23	normalizers = llm.llm.collective_rpc(lambda self: self.model_runner.model.model.normalizer.cpu().item())	CODE
CRITICAL	aphrodite/v1/spec_decode/llm_base_proposer.py	1343	self.model.model.embed_tokens.weight.cpu(),	CODE
CRITICAL	aphrodite/model_executor/layers/fla/ops/utils.py	158	triton.runtime.driver.active.utils.get_device_properties(i)["max_shared_mem"]	CODE

Modern AI Meta-Vocabulary25 hits · 72 pts

Severity	File	Line	Snippet	Context
MEDIUM	config.yaml	22	# The maximum sequence length/context window for the model	COMMENT
MEDIUM⚡	tools/pre_commit/generate_attention_backend_docs.py	1425	# Top-level orchestration	COMMENT
MEDIUM⚡	tests/v1/kv_offload/tiering/p2p/test_manager.py	1026	# _poll_once orchestration	COMMENT
MEDIUM	tests/v1/kv_connector/unit/test_offloading_connector.py	151	# Use a long prompt that fits within the model's context window.	COMMENT
MEDIUM	tests/v1/engine/test_async_llm.py	830	# Wait for generation to start (event-driven)	COMMENT
MEDIUM	tests/v1/engine/test_async_llm.py	872	# Wait for some tokens (event-driven, handles slow token generation)	COMMENT
MEDIUM	tests/v1/engine/test_async_llm.py	929	# Wait for at least one token across any request (event-driven)	COMMENT
MEDIUM	tests/evals/mrcr/mrcr_eval.py	32	# Skip chain-of-thought on reasoning models; ignored by non-reasoning templates.	COMMENT
MEDIUM	tests/compile/fullgraph/test_basic_correctness.py	57	# embedding model	COMMENT
MEDIUM	tests/entrypoints/openai/responses/test_harmony.py	584	# NOTE: chain-of-thought should be removed.	COMMENT
MEDIUM	…ntrypoints/openai/chat_completion/test_serving_chat.py	1113	# adding max_tokens should exceed the model context window.	COMMENT
MEDIUM	docs/src/content/docs/usage/tool-call.md	171	The tool calling that is supported is the [JSON-based tool calling](https://llama.meta.com/docs/model-cards-and-prompt-f	CODE
MEDIUM	aphrodite/_xpu_ops.py	813	window_size=(-1, -1), # -1 means infinite context window	CODE
MEDIUM	aphrodite/v1/attention/ops/vit_attn_wrappers.py	218	# Without it, hallucinations occur with the backend	COMMENT
MEDIUM	aphrodite/v1/executor/abstract.py	44	uses_ray: bool = False # whether the executor uses Ray for orchestration.	CODE
MEDIUM	aphrodite/metal/stt/whisper/config.py	13	# Maximum decode tokens for Whisper models (matches Whisper context window).	COMMENT
MEDIUM⚡	…ibuted/kv_transfer/kv_connector/v1/nixl/push_worker.py	243	# progress is event-driven (see module docstring).	COMMENT
MEDIUM	…d/kv_transfer/kv_connector/v1/mooncake/store/worker.py	4	# The transfer-thread scaffolding (KVTransferThread, KVCacheStoreSendingThread,	COMMENT
MEDIUM	aphrodite/model_executor/models/granite_speech.py	79	# work pretty well with zero shot.	COMMENT
MEDIUM	aphrodite/model_executor/models/phi3v.py	561	# initialized as an embedding model	COMMENT
MEDIUM	aphrodite/models/minimax_m3/amd/model.py	1099	# NVIDIA copy — it only orchestrates the shared vision tower + the per-platform	COMMENT
MEDIUM	aphrodite/reasoning/gemma4_utils.py	27	print(result["thinking"]) # chain-of-thought or None	STRING
MEDIUM	aphrodite/reasoning/gemma4_utils.py	76	>>> print(result["thinking"]) # chain-of-thought reasoning or None	STRING
MEDIUM	aphrodite/entrypoints/speech_to_text/base/serving.py	380	# which is a strong sign of hallucination in outputs.	COMMENT
MEDIUM	aphrodite/vllm_flash_attn/flash_attn_interface.py	138	window_size=(-1, -1), # -1 means infinite context window	CODE

Fake / Example Data60 hits · 70 pts

Severity	File	Line	Snippet	Context
LOW	tests/v1/attention/test_mla_backends.py	1256	["placeholder"],	CODE
LOW	tests/v1/attention/test_dspark_noncausal_sparse_mla.py	295	builder = backend_cls.get_builder_cls()(kv_cache_spec, ["placeholder"], aphrodite_config, device)	CODE
LOW	tests/v1/attention/test_sparse_mla_backends.py	413	builder = builder_cls(kv_cache_spec, ["placeholder"], aphrodite_config, device)	CODE
LOW	tests/v1/attention/test_attention_backends.py	472	["placeholder"],	CODE
LOW	tests/v1/attention/test_attention_backends.py	608	"placeholder": PerLayerParameters(	CODE
LOW	tests/v1/attention/test_attention_backends.py	652	kv_cache_spec, ["placeholder"], aphrodite_config, device	CODE
LOW⚡	tests/v1/kv_connector/unit/test_nixl_push_connector.py	674	payload = _registration_data("placeholder")	CODE
LOW	tests/tool_parsers/test_kimi_k2_tool_parser.py	82	'{"to": "user@example.com", "subject": "Daily Update"}',	CODE
LOW	tests/tool_parsers/test_kimi_k2_tool_parser.py	89	{"to": "user@example.com", "subject": "Daily Update"},	CODE
LOW	tests/tool_parsers/test_mistral_tool_parser.py	285	"name": "John Doe",	CODE
LOW	tests/tool_parsers/test_mistral_tool_parser.py	598	"name": "John Doe",	CODE
LOW	tests/tool_parsers/test_mistral_tool_parser.py	910	"name": "John Doe",	CODE
LOW	tests/tool_parsers/test_mistral_tool_parser.py	278	"""[TOOL_CALLS] [{"arguments":{"name": "John Doe"}, "name": "get_age"}]""", # noqa: E501	STRING
LOW	tests/tool_parsers/test_mistral_tool_parser.py	591	"""[TOOL_CALLS] [{"arguments": {"name": "John Doe"}, "name": "get_age"}]""", # noqa: E501	STRING
LOW	tests/tool_parsers/test_mistral_tool_parser.py	903	"""[TOOL_CALLS] [{"arguments": {"name": "John Doe"}, "name": "get_age"}]""", # noqa: E501	STRING
LOW⚡	tests/tool_parsers/test_olmo3_tool_parser.py	23	"register_user(name='John Doe', "	CODE
LOW⚡	tests/tool_parsers/test_olmo3_tool_parser.py	31	"register_user(name='John Doe', "	CODE
LOW⚡	tests/tool_parsers/test_olmo3_tool_parser.py	40	arguments='{"name": "John Doe", '	CODE
LOW⚡	tests/tool_parsers/test_pythonic_tool_parser.py	23	"register_user(name='John Doe', "	CODE
LOW⚡	tests/tool_parsers/test_pythonic_tool_parser.py	32	arguments='{"name": "John Doe", '	CODE
LOW⚡	tests/tool_parsers/test_lfm2_tool_parser.py	26	"register_user(name='John Doe', "	CODE
LOW⚡	tests/tool_parsers/test_lfm2_tool_parser.py	35	arguments='{"name": "John Doe", '	CODE
LOW	tests/tool_parsers/test_lfm2_tool_parser.py	324	"[grocery.orderIngredients(ingredientList=[{'name': 'apple', 'quantity': '2'}], deliveryAddress='123 Main St')]"	CODE
LOW⚡	tests/tool_parsers/test_hunyuan_a13b_tool_parser.py	41	'<tool_calls>[{"name": "get_weather", "arguments": {"city": "San Francisco", "metric": "celsius"}}, {"name":	CODE
LOW⚡	tests/tool_parsers/test_hunyuan_a13b_tool_parser.py	47	"name": "John Doe",	CODE
LOW	tests/kernels/helion/test_pattern_matching.py	121	input_node = next(n for n in gm.graph.nodes if n.op == "placeholder")	CODE
LOW	tests/kernels/helion/test_pattern_matching.py	186	input_node = next(n for n in gm.graph.nodes if n.op == "placeholder")	CODE
LOW	…rnels/attention/test_rocm_aiter_mla_decode_metadata.py	114	layer_name = "placeholder"	CODE
LOW⚡	…ts/models/multimodal/processing/test_audioflamingo3.py	126	dummy_data = builder.get_dummy_mm_data(100, mm_counts, {})	CODE
LOW⚡	…ts/models/multimodal/processing/test_audioflamingo3.py	128	assert "audio" in dummy_data	CODE
LOW⚡	…ts/models/multimodal/processing/test_audioflamingo3.py	129	assert len(dummy_data["audio"]) == 2	CODE
LOW⚡	…ts/models/multimodal/processing/test_audioflamingo3.py	132	assert len(dummy_data["audio"][0]) == expected_len	CODE
LOW⚡	tests/compile/test_graph_partition.py	175	assert [node.op for node in splitting_gm.graph.nodes] == ["placeholder"] + 2 * ["call_function"] + ["output"]	CODE
LOW	tests/compile/test_graph_partition.py	58	if node.op == "call_function" and node.target == operator.getitem and node.args[0].op == "placeholder":	CODE
LOW	tests/compile/test_graph_partition.py	112	if node.op == "call_function" and node.target == operator.getitem and node.args[0].op == "placeholder":	CODE
LOW	tests/compile/test_graph_partition.py	397	if node.op == "placeholder" and isinstance(node.meta.get("example_value"), torch.SymInt)	CODE
LOW	tests/compile/test_graph_partition.py	456	if n.op == "placeholder" and isinstance(n.meta.get("example_value"), torch.SymInt)	CODE
LOW	tests/compile/test_graph_partition.py	461	if n.op == "placeholder"	CODE
LOW	tests/compile/test_graph_partition.py	564	if n.op != "placeholder":	CODE
LOW⚡	tests/compile/passes/ir/test_clone_cleanup.py	225	x_node = [n for n in graph_module.graph.nodes if n.op == "placeholder"][0]	CODE
LOW⚡	tests/compile/passes/ir/test_clone_cleanup.py	237	placeholders = [n for n in graph_module.graph.nodes if n.op == "placeholder"]	CODE
LOW	tests/compile/passes/ir/test_clone_cleanup.py	253	placeholders = [n for n in graph_module.graph.nodes if n.op == "placeholder"]	CODE
LOW⚡	tests/compile/passes/ir/test_clone_cleanup.py	268	placeholders = [n for n in graph_module.graph.nodes if n.op == "placeholder"]	CODE
LOW⚡	tests/compile/passes/ir/test_clone_cleanup.py	282	x_node = [n for n in graph_module.graph.nodes if n.op == "placeholder"][0]	CODE
LOW⚡	tests/compile/passes/ir/test_clone_cleanup.py	299	x_node = [n for n in graph_module.graph.nodes if n.op == "placeholder"][0]	CODE
LOW⚡	tests/benchmarks/test_txt_slices_dataset.py	20	Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor	CODE
LOW⚡	tests/benchmarks/test_txt_slices_dataset.py	20	Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor	CODE
LOW	…hrodite/model_executor/models/transformers/fx_utils.py	53	if node.op == "placeholder" and node.target == "hidden_states":	CODE
LOW	…odite/model_executor/models/transformers/fusers/qkv.py	61	and node.args[0].op == "placeholder"	CODE
LOW	…/model_executor/models/transformers/fusers/rms_norm.py	135	x = find_node(graph, lambda n: n.op == "placeholder")	CODE
LOW	…/model_executor/models/transformers/fusers/rms_norm.py	175	if (x := find_node(graph, lambda n: n.op == "placeholder")) is None:	CODE
LOW	aphrodite/compilation/piecewise_backend.py	30	if node.op == "placeholder":	CODE
LOW	aphrodite/compilation/piecewise_backend.py	59	if node.op != "placeholder":	CODE
LOW	aphrodite/compilation/backends.py	451	input_node.op == "placeholder" or node_to_subgraph_id[input_node] <= prev_non_splitting_subgraph_id	CODE
LOW⚡	aphrodite/compilation/backends.py	532	if node.op in ("output", "placeholder"):	CODE
LOW⚡	aphrodite/compilation/backends.py	542	if input_node.op != "placeholder":	CODE
LOW	aphrodite/compilation/backends.py	1197	for i in graph.graph.find_nodes(op="placeholder"):	CODE
LOW	aphrodite/compilation/codegen.py	61	if node.op == "placeholder":	CODE
LOW	…ite/compilation/passes/ir/inplace_functionalization.py	82	if arg.op == "placeholder":	CODE
LOW⚡	aphrodite/compilation/passes/ir/clone_elimination.py	124	if original_node.op == "placeholder" and node_to_idx[original_node] not in donated_input_ids:	CODE

TODO Padding11 hits · 16 pts

Severity	File	Line	Snippet	Context
LOW	csrc/cpu/sgl-kernels/common.h	365	// TODO: implement reverse order of [MB / cache_blocks_mb, NB, cache_blocks_mb]	COMMENT
LOW	…nector/v1/lmcache_integration/multi_process_adapter.py	575	# TODO: add error handling here	COMMENT
LOW	…e/model_executor/kernels/linear/mixed_precision/xpu.py	181	# TODO: implement asym case	COMMENT
LOW	aphrodite/entrypoints/openai/responses/protocol.py	692	# TODO: implement the other reason for incomplete_details,	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	308	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	341	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	365	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	402	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	432	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	451	# TODO: implement it based on torch ops	COMMENT
LOW	aphrodite/lora/punica_wrapper/punica_base.py	478	# TODO: implement it based on torch ops	COMMENT

AI Response Leakage2 hits · 15 pts

Severity	File	Line	Snippet	Context
HIGH	tests/v1/sample/test_logprobs.py	883	# Based on user's example: "In this example,"	COMMENT
HIGH	examples/vision/vision_example.py	115	# In this example, we override max_num_seqs to 5 while	COMMENT

Example Usage Blocks8 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW	formatting.sh	4	# Usage:	COMMENT
LOW	docker/docker-bake.hcl	5	# Usage:	COMMENT
LOW	docker/export_wheels.sh	6	# Usage:	COMMENT
LOW	…/nixl_integration/run_multi_connector_accuracy_test.sh	15	# Usage:	COMMENT
LOW	…nector/nixl_integration/spec_decode_acceptance_test.sh	13	# Usage:	COMMENT
LOW	…nixl_integration/run_multi_connector_edge_case_test.sh	15	# Usage:	COMMENT
LOW	examples/run_cluster.sh	8	# Usage:	COMMENT
LOW	examples/multi_node.sh	11	# Example usage:	COMMENT

Slop Phrases5 hits · 12 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/models/multimodal/generation/test_common.py	64	# model arch happens to be a substring of another one, you can add a	COMMENT
MEDIUM	tests/models/multimodal/generation/test_common.py	85	# NOTE you can add --collect-only to any of the above commands to see	COMMENT
MEDIUM	…disagg_prefill_lmcache_v1/disagg_aphrodite_launcher.sh	24	# secure random value. This is set to a fixed value for demonstration purposes only.	COMMENT
LOW	…ibuted/kv_transfer/kv_connector/v1/nixl/pull_worker.py	79	# while processing the next batch, we make sure to only set an	COMMENT
MEDIUM	aphrodite/model_executor/models/interfaces.py	228	as a language model component.	STRING

Synthetic Comment Markers1 hit · 8 pts

Severity	File	Line	Snippet	Context
HIGH	docs/src/content/docs/prompt-caching/introduction.md	22	# A prompt containing a large markdown table. The table is randomly generated by GPT-4.	COMMENT

Overly Generic Function Names4 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW	examples/disaggregated_prefill/p2p_nccl/proxy_server.py	122	async def handle_request():	CODE
LOW	benchmarks/disagg_benchmarks/nixl_proxy_server.py	180	async def handle_request():	CODE
LOW	…marks/disagg_benchmarks/disagg_prefill_proxy_server.py	134	async def handle_request():	CODE
LOW	benchmarks/disagg_benchmarks/round_robin_proxy.py	15	async def handle_request(self, request):	CODE

Dead Code1 hit · 2 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/v1/e2e/general/test_streaming_input.py	473		CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings