vllm-project/vllm

main

vllm

Grade	Name	Complexity	Churn	Issues
A	vllm\entrypoints\pooling\base\io_processor.py	25	2	1
A	benchmarks\kernels\benchmark_cutlass_moe_fp8.py	25	1	-
A	tests\distributed\test_pynccl.py	25	1	-
A	tests\distributed\test_quick_all_reduce.py	25	1	-
A	vllm\model_executor\models\blip2.py	24	1	1
A	vllm\model_executor\models\longcat_flash_ngram.py	24	1	-
A	benchmarks\kernels\benchmark_w8a8_block_fp8.py	24	1	2
A	tests\kernels\quantization\test_marlin_tile_padding.py	24	1	-
A	vllm\reasoning\olmo3_reasoning_parser.py	24	1	-
A	vllm\model_executor\models\step1.py	24	1	-
A	vllm\model_executor\parameter.py	24	1	3
A	csrc\cpu\sgl-kernels\moe_fp8.cpp	24	1	-
A	tests\v1\core\test_async_scheduler.py	24	2	-
A	vllm\model_executor\models\exaone_moe.py	24	1	3
A	vllm\entrypoints\scale_out\derender\serving.py	24	1	1
A	vllm\model_executor\layers\pooler\tokwise\heads.py	24	1	1
A	vllm\distributed\kv_transfer\kv_connector\v1\nixl\push_scheduler.py	24	1	-
A	vllm\model_executor\models\modernbert.py	24	1	-
A	vllm\model_executor\models\qwen3_5.py	24	1	1
A	vllm\model_executor\models\jais2.py	24	1	3