Unverified50% confidenceBenchmarkExact time
At approximately 4K input, TTFT is 23.6 seconds with decoding at 46.6 tok/s; at 8K, 47 seconds and 41.4 tok/s; at 16K, 1 minute 30 seconds and 30.8 tok/s; at 30K, 3 minutes 15 seconds and 31.9 tok/s.
1
Sources
50%
Confidence
Long-term
Relevance
10/2/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedOn a T4 GPU, Laya takes 33 milliseconds per single question, and batch processing can compress that to 7 milliseconds79% similarUnverifiedThe cumulative latency of a traditional ASR→LLM→TTS three-stage pipeline is typically between 800 milliseconds and 2 seconds.78% similarUnverifiedOn a 16G GPU, a 60-second audio clip takes only 101 seconds with 512-frame chunking, 231 seconds with 1024-frame chunking, and 285 seconds without chunking.75% similarUnverified一个需要20次LLM调用、每次生成200个Token的Agent任务，在30 TPS下需要约133秒完成，在300 TPS下仅需13秒75% similarUnverified每条DNS记录都附带TTL值，指定递归解析器可缓存该记录的秒数，常见设置从300秒到86400秒不等75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/968806API
curl https://kongchang.com/api/v1/knowledge/claims/968806MCP
get_claim(id=968806)