About this paper
AscendGYM provides a benchmark for assessing kernel optimization capabilities on Ascend NPUs using real-workload test cases, two evaluation modes, and a Speedup-Token Score that balances speedup against token cost.
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 ability of agents to optimize existing implementations under input configurations drawn from real workloads. To address this requirement, we introduce AscendGYM, a benchmark whose initial suite comprises 48 kernels across three difficulty levels. Each kernel is paired with test cases whose data types and tensor shapes are collected from real-world training and inference tasks. Consequently, performance improvements on AscendGYM map directly to the corresponding workloads. In terms…
large language model (LLM)Ascend NPUskernel optimizationkernel benchmarkskernel generationagentsinput configurationstensor shapesmodel evaluationmodel-harness evaluation
- Upvotes
- 83
- Authors
- 27
- Published
- Nov 13, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
AscendGYM: A Cost-Aware Kernel Optimization Benchmark for Real-World Workloads on Ascend NPUs was first seen on Hugging Face · Daily papers on Nov 18, 2026. Since then it has appeared in 12 checks on 2 different days. Its best position was #2 of 30, reached on Nov 18, 15:28 UTC. It dropped off the list after Nov 19, 05:28 UTC.