DeepSeek V4.1-Flash open-sourced: CED architecture cuts prefill cost for coding agents

DeepSeek 开源 V4.1-Flash：CED 架构降低编码 Agent 的 prefill 开销

DeepSeek released open weights for V4.1-Flash, a 552B MoE model with a Causal Encoder-Decoder architecture tuned for coding agents. It splits compute asymmetrically: 8B active params during prefill, 16B during decode, plus improved KV cache efficiency. On Terminal Bench 2.1 it hits 90.6; on Automation-Bench it scores 54.8—better than V4-Pro but still failing roughly half of complex workflows, so keep a human in the loop. It is also DeepSeek's first non-experimental model with native image input. Chartography reaches 78.9, but ZeroBench logical reasoning over images is only 49. DeepSeek has already retired V4-Flash traffic and will reroute V4-Pro traffic to V4.1-Flash starting September 14.

Why it matters: DeepSeek open-sourced V4.1-Flash, a 552B MoE that splits prefill and decode via CED architecture, directly targeting coding agent latency. Terminal Bench 2.1 scores are concrete, and Baseten's analysis adds deployment perspective. Not 85+ because this is a third-party writeup ...