DeepSeek Open Sources DeepInfer Under Apache 2.0: Production Serving Engine for NVIDIA SM100 and Huawei Ascend 950
software••2 min•github.com
DeepSeek has open-sourced its production-grade offline inference runtime, DeepInfer, under the Apache 2.0 license. The system features FlashMLA, DeepGEMM FP8 kernels, and DSpark speculative decoding with native support for both NVIDIA SM100 and Huawei Ascend 950 silicon.
Read full article→#DeepSeek