Unverified50% confidenceBenchmarkExact time
On Jetson AGX Thor, d1-3B takes 16 milliseconds to answer one question and 20 milliseconds for three questions; on Jetson AGX Orin 64GB it is 26 milliseconds and 35 milliseconds; on Jetson Orin Nano it is 50 milliseconds and 73 milliseconds.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
VerifiedOn a T4 GPU, Laya takes 33 milliseconds per single question, and batch processing can compress that to 7 milliseconds69% similarUnverifiedAt 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.67% 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.67% similarUnverifiedTraditional Electron applications typically require hundreds of milliseconds to several seconds for a cold start, with a memory baseline of 100–300 MB.66% similarUnverifiedTested on the same machine equipped with a Spark CPU and limited to three threads, Laya's median call latency was 541 milliseconds and Kev's was 756 milliseconds, with the 95th percentile at approximately 607 and 802 milliseconds respectively.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013596API
curl https://kongchang.com/api/v1/knowledge/claims/1013596MCP
get_claim(id=1013596)