FlashAttention 怎么读

FlashAttention 在英语中的标准读法是 「flash uh ten shun」（/ˌflæʃ əˈtɛnʃən/）。

IO-aware exact-attention GPU kernel (Dao-AILab; FlashAttention-2/3) baked into vLLM, TGI, TensorRT-LLM, PyTorch SDPA. Spoken as the two English words "flash" + "attention" run together, one word in writing. STRESS: secondary on FLASH, primary on -TEN- (flash-uh-TEN-shun). Non-native speakers hesitate on the flash+attention compound; pronunciation is just the plain English words.

来源：arXiv 2205.14135 — FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness (Dao et al.)

读对项目名只是工程师文化的一部分 — 但开会时被人轻轻纠正、或者在 podcast 上被嘉宾礼貌打断的滋味，并不好受。Pronounce 是一个开源的开发者词汇发音词典，每条都标注来源、信心等级，并且有预渲染的音频可以直接听。