Nvidia Tilus: A Tile-Level GPU Kernel Programming Language
10 days ago
- #Domain-Specific Language
- #GPU programming
- #Low-Precision Computation
- Tilus is a domain-specific language (DSL) for GPU programming with thread-block-level granularity and tensors as the primary data type.
- It offers explicit control over shared memory and register tensors, unlike Triton, and supports low-precision types with arbitrary bit-widths (1 to 8 bits).
- Features include automatic tuning, caching, and a Pythonic interface for ease of use.
- Pronounced as tie-lus, /��ta��ləs/, and can be installed via pip: `pip install tilus`.
- Based on the research paper titled 'Tilus: A Virtual Machine for Arbitrary Low-Precision GPGPU Computation in LLM Serving'.
- Acknowledges influence from projects like Hidet, TVM, Triton, and Hexcute in its design and development.