Unverified50% confidenceBenchmarkExact time
在实际基准测试中，FTS5 对百万级文档的检索延迟通常在个位数毫秒级别
1
Sources
50%
Confidence
Long-term
Relevance
8/26/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn testing on an RTX 5080, the BF16 baseline took approximately 4.651 seconds per sample, while the selective FP8 approach took approximately 3.651 seconds, a speedup of approximately 22%.71% similarUnverifiedThe time-to-first-byte for cloud TTS services is typically 200–500 milliseconds, making it the primary source of latency bottlenecks.70% similarUnverifiedHNSW在百万量级向量库中的查询延迟通常仅需1-10毫秒，召回率可达99%以上69% similarUnverified自回归token生成的延迟通常在数百毫秒量级，难以满足实时控制需求68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/804543API
curl https://kongchang.com/api/v1/knowledge/claims/804543MCP
get_claim(id=804543)