Unverified50% confidenceBenchmarkExact time
Phi-3 Mini（3.8B参数）在多项推理基准上超越了早期的GPT-3.5，且体积小到可在普通智能手机上运行
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/17/2026
First Seen
Valid until: 10/15/2026
Sources
Related Claims
Unverified微软的Phi-3仅3.8B参数但在多项基准测试中性能媲美早期GPT-4级别的模型77% similarUnverifiedPhi-3-mini（3.8B参数）在部分基准测试上超越了Mixtral 8x7B等更大的混合专家模型70% similarUnverified微软的Phi-3 Mini仅有38亿参数，但在多项基准测试中超越了参数量数倍于它的模型65% similarUnverifiedMicrosoft的Phi-3-mini（3.8B参数）在多项基准测试中超越了早期的7B-13B模型64% similarUnverifiedMiniMax 2.7的定价远低于GPT-4或Claude 3.5等顶级模型64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/541303API
curl https://kongchang.com/api/v1/knowledge/claims/541303MCP
get_claim(id=541303)