Open Source · DeepSeek · Nvidia · China · Huawei · Crypto Briefing
DeepSeek open-sources DeepInfer, the inference engine behind its production models
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
◌ Single Source
The Chinese AI lab released its high-performance serving framework under Apache-2.0, alongside a token encoding tool called deepseek-recipe.
Key facts
- The release landed on October 20, 2026, under the Apache-2.0 license
- Supporting both NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs puts a Chinese chip on equal footing with NVIDIA in an official release
- Generation speed runs at 230 tokens per second, reaching up to 300 tokens per second for coding tasks
- The Chinese AI lab released its high-performance serving framework under Apache-2.0, alongside a token encoding tool called deepseek-recipe
Summary
DeepSeek has open-sourced DeepInfer, the high-performance inference framework that runs its own models in production. The release landed on October 20, 2026, under the Apache-2.0 license. DeepInfer bundles a set of optimizations aimed at making inference faster and cheaper to run. Custom kernels are low-level routines written to squeeze more work out of specific chips.