{"id":"d2f5a46e-47c1-43c9-a2fe-2afa16c8e1f9","url":"https://api.tumn.com/news/d2f5a46e-47c1-43c9-a2fe-2afa16c8e1f9","title":"Accelerating Mamba-2 with High-Performance Forward Pass Kernels in NVIDIA CUDA Tile","summary":"NVIDIA CUDA Tile (cuTile) is a Python-based DSL and parallel programming model for NVIDIA GPUs that automatically leverages advanced hardware capabilities to...","sections":[],"faq":[],"sources":[{"url":"https://developer.nvidia.com/blog/accelerating-mamba-2-with-high-performance-forward-pass-kernels-in-nvidia-cuda-tile/","title":"NVIDIA Developer Blog","publishedAt":"2026-10-19T17:00:00.000Z"}],"author":"TUMN 큐레이션","aiAssisted":false,"reviewStatus":"automated","reviewNote":"원문 제목과 공개된 발췌를 인용하고 원문으로 연결합니다. 내용을 재작성하거나 요약하지 않았습니다.","publishedAt":"2026-10-19T17:00:00.000Z","modifiedAt":"2026-10-19T20:25:34.000Z","checkedAt":"2026-10-19T20:25:34.302Z","language":"ko","kind":"link","sourceLabel":"AI 공식","profile":null,"via":null,"formats":{"json":"https://api.tumn.com/news/d2f5a46e-47c1-43c9-a2fe-2afa16c8e1f9.json","markdown":"https://api.tumn.com/news/d2f5a46e-47c1-43c9-a2fe-2afa16c8e1f9.md"}}