DeepInfer: High-performance inference framework for DeepSeek models

DeepInfer is an open-source, high-performance inference framework designed for DeepSeek models, enabling large-scale production services. It supports offline inference with features such as optimized kernels, parallelism, asynchronous scheduling, and deterministic sampling. The framework is compatible with NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs, and it allows for features like batch-invariant inference and structured outputs.