Verified65% confidenceFactTime unknown
中间层轻量模型（如Claude 3 Haiku、GPT-4o-mini）价格约为旗舰模型的1/10至1/20
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
ccusage：一行命令算清AI编程真实花费
bilibili锋芒AI
Related Entities
Related Claims
UnverifiedMiniMax 2.7的定价远低于GPT-4或Claude 3.5等顶级模型69% similarUnverifiedClaude 3 Haiku的定价仅为Claude 3 Opus的约1/6065% similarUnverifiedClaude Haiku 4.5在三个前端测试场景中生成速度最快，成本最低62% similarUnverified一些开源或轻量级模型（如Mistral、Llama系列通过第三方托管）的调用成本可能只有Claude 3.5 Sonnet的十分之一甚至更低61% similarUnverified在DeepSWE上，各模型分数从顶部GPT 5.5的70%到底部Cloud Haiku 4.5的0%，区分度远大于SWE-Bench Pro约30分的集中区间60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/9825API
curl https://kongchang.com/api/v1/knowledge/claims/9825MCP
get_claim(id=9825)