这篇论文展示了如何在华为Ascend芯片上高效后训练万亿参数模型，并且通过运筹学专用微调，在特定任务上超越了GPT-5.4-Mini。
该论文针对万亿参数MoE模型DeepSeek-V4在Ascend NPU SuperPOD上的全参数后训练，提出分层优化框架，实现34.22%的模型算力利用率（MFU），较开源基线提升2.93倍。基于优化基础设施，论文构建了面向运筹学（OR）任务的连续预训练（CPT）和微调（SFT）工作流。SFT数据集包含10K高质量样本，涵盖4个任务类别和3种问题表示。专用模型DeepSeek-V4-Flash-OR在零样本Pass@1上达到71.81%，比GPT-5.4-Mini高3.98个百分点，比基础DeepSeek-V4-Flash高11.27个百分点。
SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD
Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pressure, non-overlapped communication overhead, and inefficient kernel execution. While most large-scale LLM training systems are built around GPU-based clusters, this report presents an end-to-end optimization practice on the Ascend NPU SuperPOD. Using the DeepSeek-V4 model family as the target workload, we develop a hierarchical optimization framework spanning model-level parallelism, computation-communication orchestration, and low-level kernel execution. The resulting system achieves 34.22% Model FLOPs Utilization (MFU) with a 2.93x improvement over the open-source baseline recipe while maintaining training stability. Building on this optimized infrastructure, we further establish a CPT and SFT workflow for complex Operations Research (OR) tasks. We refer to the integrated framework as SLAI T-Rex. Using DeepSeek-V4-Flash, we develop OR-oriented CPT and SFT data pipelines that combine collected domain resources with solver-verified synthetic optimization documents. The resulting dataset contains 10K high-quality SFT samples spanning four task categories and three problem representations. The specialized model achieves the highest average zero-shot Pass@1 score among the evaluated models, reaching 71.81% and outperforming GPT-5.4-Mini and the base DeepSeek-V4-Flash model by 3.98 and 11.27 percentage points, respectively. Overall, this work demonstrates a full-stack pathway from efficient trillion-parameter model post-training on Ascend infra to domain-specialized Flash models for solver-grounded mathematical modeling, advancing frontier-model systems for complex reasoning.