Unverified50% confidenceBenchmarkExact time
Running Kimi K2.5 on an NVIDIA B200 with real coding agent trajectories (SWE-smith, EvalScope evaluation), TokenSpeed outperforms TensorRT-LLM across the entire latency/throughput curve; at batch size 1 it is about 9% faster, and at about 100 tokens/second per user its throughput is about 11% higher.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/15/2026
First Seen
Valid until: 10/29/2026
Sources
Related Entities
Related Claims
UnverifiedJev's reported server time is approximately 88 milliseconds per classification, while running GLiNER 2.5 small locally on a MacBook M4 Pro GPU is approximately 8 times faster.70% similarUnverified根据用户反馈，使用 Quantprobe 的深度感知优化配方，相比默认设置生成速度可以几乎翻倍，在 GTX 1060 上从约 10 tokens/s 跃升至超过 22 tokens/s70% similarUnverifiedIn the test, model loading took only a few seconds, GPU usage dropped from 23GB to about 17GB, the first test prompt processing took 225ms, generation speed was stable at 33.5 tok/s, with a total output of 1500 tokens.70% similarUnverifiedThe typical generation speed of an NVIDIA RTX 4090 running a similarly scaled Q4 quantized model is in the range of 40–60 TPS69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1041082API
curl https://kongchang.com/api/v1/knowledge/claims/1041082MCP
get_claim(id=1041082)