Unverified50% confidenceFactExact time
DeepSeek-Infer supports limited replay with a sliding window, replaying only a small recent stretch of tokens when restoring state to reduce redundant computation.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSome of DeepSeek's model layers use sliding window attention, attending only to the most recent 128 tokens; SWA bounded replay replays only a short window at the end of the prompt, which is said to nearly eliminate total prefill work.72% similarUnverifiedWhen a model makes repeated calls with exactly the same parameters, DeepSeek Harness adds a reminder on the 3rd, 5th, and 8th occurrences prompting the model to try a different approach or stop, but does not forcibly prevent it.64% similarUnverifiedDeepSeek's official pay-as-you-go channel supports prompt caching, and content that hits the cache is not billed again in subsequent steps.63% similarUnverifiedDeepInfer uses SWA bounded replay technology to reduce the last 2,560 tokens recomputed when rebuilding the cache to 128, cutting total prefill computation nearly in half and decoupling long-context cost from length.63% similarUnverifiedAfter a DeepSeek Harness task is killed mid-execution, it can precisely resume from the point of interruption based on persisted state.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1088525API
curl https://kongchang.com/api/v1/knowledge/claims/1088525MCP
get_claim(id=1088525)