Skip to content
AIBRAINWORX / THE INDEPENDENT AI LABRESEARCH · HARDWARE · OPEN INTELLIGENCE
AI News / Open Source

DeepSeek open-sources DeepInfer under Apache 2.0 for Blackwell and Ascend

The offline inference engine pairs custom FlashMLA, DeepGEMM, and DeepEP kernels with native support for NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs.

Official DeepSeek profile mark
DeepSeek, published Hugging Face profile icon; identification only, no endorsement.
Event / context: DeepSeek open-sourced DeepInfer in November 2026.. Source check: October 28, 2026.

DeepSeek has open-sourced DeepInfer under the Apache 2.0 license, releasing the core inference engine used behind its production services. Rather than tying its runtime to a single accelerator vendor, DeepInfer ships with native support for both NVIDIA SM100 (Blackwell B200/GB200) GPUs and Huawei Ascend 950 NPUs.

Targeted kernels and batch scope

The initial release targets DeepSeek-V4.1-Flash Mixture-of-Experts serving, combining FlashMLA, DeepGEMM, DeepEP, asynchronous batch scheduling, grammar-constrained structured outputs, and DSpark speculative decoding.

What is not included out of the box

DeepInfer is positioned as an offline batch inference engine and reference implementation rather than a turnkey online microservice. It does not bundle a full HTTP serving layer, distributed KV-cache disaggregation, or replica load balancing, leaving production web-serving orchestration to downstream teams. AiBrainWorX has not independently benchmarked DeepInfer on Blackwell or Ascend hardware.

Original reporting; repository and corroborating sources reviewed November 16, 2026. No independent hardware benchmark claimed.

Follow your curiosity.

Pass the signal.