Unverified60% confidenceFactExact time
V4.1 Flash reduces KVCache to one-quarter of the previous generation.
2
Sources
60%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedV4.1 Flash scored 74.2 on the SWE software engineering benchmark, 90.6 on the terminal benchmark, and 3,471 points on Tonka Forces evaluation (vs. 3,348 for VS Pro).69% similarUnverifiedV4 Flash and V4 Flash Vision XP have been taken offline, with the old model names temporarily redirected to the new model.67% similarUnverifiedA developer claims to have partially reproduced on the Qwen model the KV cache mechanism used by DeepSeek V4.1 Flash to accelerate the prefill stage.67% similarUnverifiedAccording to official charts, V4.1 Flash's DPSAD score is 74.2.67% similarUnverifiedFlash models typically use GQA or MQA technology to compress the KV Cache by 4–8×, thereby supporting higher concurrency on the same hardware.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/908261API
curl https://kongchang.com/api/v1/knowledge/claims/908261MCP
get_claim(id=908261)