#ascend
4
2026-10-01
deepseek-ai/DeepGEMM-Ascend
S
DeepGEMM-Ascend: clean and efficient matrix multiplication kernel library for Huawei Ascend NPUs
2026-10-01
deepseek-ai/DeepEP-Ascend
A
A high-performance communication library for machine learning training and inference on Huawei Ascend NPUs.
2026-10-16
Gxj230958/TritonAscendBench
未评级
A migration benchmark that measures whether production GPU Triton kernels — extracted from vLLM and SGLang inference paths — can be ported to the Huawei Ascend 910B (Atlas A2) via `triton-ascend` without losing correctness or performance.