vllm-project/vllm

main

vllm

Grade	Name	Complexity	Churn	Issues
A	vllm\v1\sample\thinking_budget_state.py	122	1	2
A	vllm\distributed\kv_transfer\kv_connector\v1\lmcache_integration\vllm_v1_adapter.py	122	1	3
A	vllm\model_executor\models\qwen2_5_omni_thinker.py	122	1	7
C	vllm\models\deepseek_v4\nvidia\model.py	120	2	8
A	vllm\v1\executor\multiproc_executor.py	120	1	1
A	vllm\model_executor\model_loader\bitsandbytes_loader.py	120	1	2
A	benchmarks\kernels\benchmark_moe.py	119	1	4
A	csrc\cpu\cpu_types_vsx.hpp	118	1	4
A	vllm\v1\worker\gpu_worker.py	118	1	1
A	vllm\model_executor\layers\quantization\humming.py	117	1	-
A	vllm\transformers_utils\processors\mimo_v2_omni.py	117	1	2
A	csrc\cpu\sgl-kernels\fla.cpp	116	1	2
A	vllm\model_executor\layers\fused_moe\oracle\int_wna16.py	115	1	2
A	vllm\parser\gemma4.py	115	1	2
A	vllm\v1\attention\ops\rocm_aiter_mla_sparse.py	115	1	4
A	csrc\cpu\cpu_types_vxe.hpp	115	1	4
A	vllm\distributed\device_communicators\cuda_communicator.py	115	1	3
A	vllm\entrypoints\openai\chat_completion\protocol.py	114	1	7
A	benchmarks\kernels\benchmark_fused_collective.py	114	1	2
A	vllm\model_executor\layers\quantization\quark\quark_moe.py	112	1	4