Expired50% confidenceFactExact time
The researcher empirically measured the quota deduction rate of the old Pro plan by repeatedly calling the /api/usage endpoint before the billing plan switch, with testing covering August 28 to 31 and a margin of error controlled to ±7–13%
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/11/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/25/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedAccording to the user's test data, the actual resources provided by the new plan are approximately 3 to 6 times less than the previous quota, with the specific magnitude depending on the model used.71% similarUnverifiedUsers have reported that Cursor's model response speed has recently slowed down and output quality has declined.69% similarUnverifiedFastEmbed-rs所使用的量化模型在MTEB基准测试中相比全精度模型的性能下降通常控制在1-2%以内66% similarUnverifiedIn the experiment, after prompts exceeded 50, the human interception rate dropped sharply from 17% to 5%, reflecting the phenomenon of approval fatigue.65% similarUnverified将FP32权重降至INT8后，典型Transformer模型基准测试分数下降通常不超过1%，推理速度提升2至4倍，内存占用减少75%64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/899375API
curl https://kongchang.com/api/v1/knowledge/claims/899375MCP
get_claim(id=899375)