deepseek-v4.1POST
DeepSeek staff acknowledge third-party vLLM serving of V4.1 Flash that reproduced results
DeepSeek's yifandotqiao thanked a third-party blog for serving V4.1 Flash with vLLM and reproducing their results, and acknowledged the DeepInfer Engram RDMA fetch bottleneck the blog identified.
Yifan Qiao
@yifandotqiaoArchived 3 days agoOriginally posted 2026.10.25
Yifan Qiao @yifandotqiaoOriginally posted 2026.10.25
Nice writeup! So glad to see a third party serving DeepSeek-V4.1-Flash with vLLM and reproducing our results! Also love how the blog identifies DeepInfer's Engram RDMA fetch bottleneck, which we've observed and taken seriously as well. Regardless, vLLM is the full-fledged serving engine that takes care of the heavy lifting. Talk is cheap. Try it yourself!
This post quotes another post: x.com/kbesplv/status/2112913887817355463
x.com/yifandotqiao/status/2114127241068331180Archived 2026.10.25 12:00