DeepSeek Open-Sources DeepInfer; vLLM Reports Matching Its V4.1-Flash Decode Performance
DeepSeek has open-sourced DeepInfer under Apache 2.0, releasing the inference framework it says powers DeepSeek-V4.1-Flash in production; vLLM reports matching its performance on DeepSeek’s decode configuration. That comparison covers batch sizes 8 to 256, with vLLM running at DEP32 against DeepInfer’s EP64DP64. vLLM said DeepInfer was originally built primarily on vLLM. According to Baoyu, DeepInfer currently supports only V4.1-Flash on Nvidia Blackwell SM100 or Huawei Asce…
Latest developmentHow DeepInfer Coordinates Asynchronous Scheduling for DeepSeek-V4.1-Flash Inference
