Open Issues Need Help
View All on GitHub [Perf baseline] cann-recipes-infer DeepSeek-V4-Flash on Ascend 950PR (8p): full-network + per-operator numbers, and what it took to run about 2 months ago
enhancement help wanted
hw-native-sys/pypto-lib hw-native-sys/pypto-lib Tensor-level kernels and end-to-end LLM model implementations built on the PyPTO programming framework
Open project →
12
Tensor-level kernels and end-to-end LLM model implementations built on the PyPTO programming framework
Python