DeepSeek-V4.1-Flash lands on SiliconFlow, a 552B MoE with 1M context window

DeepSeek-V4.1-Flash 上线 SiliconFlow，552B MoE 支持 1M 上下文

SiliconFlow launched DeepSeek-V4.1-Flash on Day 0. It's a 552B MoE model with ~8B active params during prefill and ~16B during decode, native vision, and a 1M-token context window. KV cache footprint is about 1/4 of V4 Flash, which helps with deployment cost. MIT license keeps commercial use straightforward.

Why it matters: Same-day availability of DeepSeek V4.1-Flash on SiliconFlow, with KV cache reduced to 1/4 of V4 Flash — a clear deployment cost signal. Score held at 78 because this is a platform availability announcement; no benchmarks or real-world performance data yet.