Firehose

22 JUL 2026 · Paper

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD

This paper optimizes the training of massive neural networks on a special-purpose hardware, the Ascend SuperPOD, to improve performance and stability. Practitioners in AI/ML model training might care about the techniques and results presented here for large-scale model training on non-GPU hardware.

large-scale distributed training MoE models NPU model-level parallelism computation-communication orchestration low-level kernel execution

Browse by tag

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD