Unverified50% confidenceOpinionExact time
It is widely speculated in the industry that Flash's parameter scale is approximately one-third to one-half that of Pro.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/30/2026
First Seen
Valid until: 11/28/2026
Sources
Related Entities
Related Claims
UnverifiedThe per-token inference cost of Flash models is typically one order of magnitude lower than that of Pro-level models.66% similarUnverified单条推理通常只能占用GPU SM利用率的5%～15%，而合理的批次大小可将利用率提升至60%～90%65% similarUnverifiedInference compute is expected to account for two-thirds of total AI computing power.55% similarUnverified参数量在1亿以下的轻量模型经领域数据微调后可达95%以上意图分类准确率，单次推理延迟控制在5-20ms以内55% similarUnverifiedThe monthly quota utilization rate of Cursor Pro accounts is typically only 30–40%, and pooled scheduling can raise overall utilization to above 70–80%.54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/824198API
curl https://kongchang.com/api/v1/knowledge/claims/824198MCP
get_claim(id=824198)