vllm-project/vllm

main

vllm

Grade	Name	Complexity	Churn	Issues
A	vllm\v1\spec_decode\ngram_proposer.py	24	1	-
A	vllm\v1\sample\logits_processor\__init__.py	24	1	-
A	vllm\v1\kv_offload\cpu\policies\arc.py	24	1	1
A	vllm\model_executor\parameter.py	24	1	3
A	vllm\model_executor\models\exaone_moe.py	24	1	3
A	vllm\model_executor\layers\pooler\tokwise\heads.py	24	1	1
A	vllm\model_executor\layers\fused_moe\prepare_finalize\deepep_ht.py	24	1	3
A	tools\profiler\nsys_profile_tools\gputrc2graph.py	24	1	-
A	tests\distributed\test_pipeline_parallel.py	24	2	2
A	vllm\config\multimodal.py	23	1	-
A	tools\generate_cmake_presets.py	23	1	1
A	benchmarks\kernels\ir\bench_ir_ops.py	23	1	-
A	vllm\model_executor\models\intern_vit.py	23	1	1
A	vllm\distributed\kv_transfer\kv_connector\v1\hf3fs\utils\gather_scatter_helper.py	23	1	-
A	vllm\models\minimax_m3\amd\mtp.py	23	1	2
A	vllm\models\deepseek_v4\nvidia\ops\fused_indexer_q_cutedsl.py	23	1	-
A	vllm\tokenizers\registry.py	23	2	-
A	vllm\model_executor\models\deepencoder.py	23	1	1
A	vllm\reasoning\abs_reasoning_parsers.py	23	1	-
A	vllm\lora\layers\fused_moe.py	23	1	-