[torch.compile] Add torch inductor pass for fusing silu_and_mul with subsequent scaled_fp8_quant operations - #10867

[torch.compile] Add torch inductor pass for fusing silu_and_mul with subsequent scaled_fp8_quant operations#10867

vllm-bot merged 48 commits into

vllm-project:mainvllm-project/vllm:mainfrom

neuralmagic:sage/silu-mul-quantneuralmagic/vllm:sage/silu-mul-quantCopy head branch name to clipboard