TileLang: Bridge Programmability and Performance in Modern Neural Kernels

Modern AI algorithms increasingly adopt fused kernels for performance, but implementing them remains complex due to the lack of fine-grained control in existing compilers like Triton. We introduce TileLang, a controllable programming system for fused neural kernels. TileLang provides explicit tile-level primitives for memory placement, data movement, and parallel scheduling. To guide developers in hardware-aware programming, the TileLang introduces two key techniques: tile inference which models tile programs as fused graphs and automatically deduces tile configuration from partial annotations; and tile recommendation that suggests efficient tile configurations based on hardware profiles and heuristics. TileLang makes it easy to express a wide range of fused attention kernels in under 80 lines of Python code, reducing code size by up to 90% compared to manual implementations. Evaluations show that TileLang achieves up to 5x speedup over Triton on NVIDIA H100 and up to 6 on AMD GPUs, demonstrating its ability to bridge programmability and performance.