Skip to main content
Run any Skill in Manus
with one click

triton-ascend-optimization

Stars255
Forks48
UpdatedApril 19, 2026 at 13:47

Triton Ascend 性能优化通用策略: BLOCK_SIZE 选择 (1024-2048 for elementwise, must be <65536), grid configuration (use VEC_CORE_NUM / CUBE_CORE_NUM, 2D/3D grid for matmul / conv / reduce, 1D grid + inner loop for elementwise / pointwise), 256B alignment for memory transfers, autotune block-size patterns, fp16 / fp32 precision conversion. Bind via keywords like matmul, elementwise, reduce, block_size, grid, autotune, alignment, fp16, fp32, tile, interleaved-loop, cube-core, vec-core.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

SKILL.md
readonly