DeepSeek open-sources DeepInfer software to curb AI deployment costs

DeepSeek has open-sourced DeepInfer, the internal inference framework used to run its AI models in production. Announced on X and published on GitHub under the Apache 2.0 licence, the software aims to curb AI deployment costs. Currently, DeepInfer exclusively supports DeepSeek-V4.1-Flash. This release allows the broader community to access the engine that powers DeepSeek's production environment, providing a transparent tool for running high-performance models while optimizing the financial resources required for large-scale AI implementation.