Story
arxiv_cs_lg ยท Nov 13, 2026 ยท paper
Source brief
AscendGYM: A Cost-Aware Kernel Optimization Benchmark for Real-World Workloads on Ascend NPUs
arxiv.orgNov 13, 2026
original source linked
In brief
Kernel optimization is essential for efficient large language model (LLM) training and inference on Ascend NPUs. Existing Ascend kernel benchmarks primarily focus on kernel generation, with less emphasis on the abilit...
Feed lens
agentharnessevaluation