Unverified50% confidenceFactExact time
V4.1 Flash uses CSA2 (Compressed Sparse Attention 2) technology combined with FP4 format for caching data, compressing the global KV Cache footprint to only 890 bytes per token.
1
Sources
50%
Confidence
Long-term
Relevance
9/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDeepSeek V4.1 Flash reduces per-token storage from 3,500 bytes in the previous generation to approximately 890 bytes through KV cache inter-layer alternation and reuse optimization.79% similarUnverifiedAs the context length increases, the rate of memory usage growth for V4.1 Flash becomes increasingly gradual, allowing it to support longer input windows with fewer resources.74% similarUnverifiedFlash Attention reduces the number of reads and writes to GPU high-bandwidth memory by reorganizing the computation order, achieving a 2–4× speed improvement without changing the computation results.72% similarVerifiedQuantized KV caches (such as storing KV vectors in FP8 or INT4) can compress the cache size to 1/2 or even 1/4 of the original.72% similarUnverifiedIn DeepSeek-V4.1-Flash, only four layers are responsible for writing the compressed KV, and the rest of the model shares this cache.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/941026API
curl https://kongchang.com/api/v1/knowledge/claims/941026MCP
get_claim(id=941026)