Unverified50% confidenceBenchmarkExact time
Phi-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
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedMicrosoft's Phi-3-mini (3.8B parameters) surpassed earlier 7B–13B models on multiple benchmark tests.74% similarUnverifiedPhi-3 Mini (3.8B parameters) outperforms early GPT-3.5 on multiple reasoning benchmarks and is small enough to run on an ordinary smartphone70% similarUnverifiedQwen3's 235B-parameter MoE flagship model approaches or surpasses closed-source top-tier models on multiple authoritative benchmarks.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/525475API
curl https://kongchang.com/api/v1/knowledge/claims/525475MCP
get_claim(id=525475)