Unverified50% confidenceBenchmarkExact time
Tether's multi-hop data-flow path traversal query latency is under 2 milliseconds.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedIn the email classification task, TEF 1 (4B) took 1.273 seconds to run through all samples, while GPT needed 2.568 seconds; in the sponsorship inquiry scoring task, TEF 1 took 1.09 seconds while GPT took about 2.58 seconds.70% similarUnverifiedtps (tokens per second) is a core metric for measuring model inference speed: below 30 tps you will clearly feel the wait, 60-100 tps can satisfy most dialogue and coding needs, and above 100 tps is quite smooth.70% similarUnverifiedA Time to First Token (TTFT) below 500ms is generally considered a smooth interactive experience, while exceeding 2 seconds creates a noticeable sense of waiting.69% similarUnverifiedCompared to pure encoder inference, the additional latency introduced by RATED is less than 1 millisecond.69% similarUnverified缓存命中时prefill阶段可跳过已缓存部分，TTFT可从数秒压缩至毫秒级69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1062002API
curl https://kongchang.com/api/v1/knowledge/claims/1062002MCP
get_claim(id=1062002)