Expired50% confidenceFactExact time
The MiniCPM series achieves performance close to 7B and even 13B models on multiple benchmarks with a parameter count at the 2B–4B level
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/30/2026(expired)
Sources
PilotDeck: A Local Console That Tames Multi-Task Agent Chaos
bilibiliAIlazy俊6/4/2026
Related Claims
UnverifiedMicrosoft's Phi-3-mini (3.8B parameters) surpassed earlier 7B–13B models on multiple benchmark tests.73% similarUnverified在27B混合架构、4B稠密模型、36B MoE三类模型上以8k上下文测试，Prefill阶段性能无变化，Decode阶段打补丁后快约1.4%71% similarUnverified万象模型提供了 1.3B 和 14B 参数量的不同规模版本70% similarVerifiedThe larger the number of model parameters, the greater the robustness to quantization, allowing 70B-level models to maintain performance close to FP16 after 4-bit quantization.70% similarUnverifiedMistral 7B证明精心训练的7B模型可以匹敌早期的13B甚至30B模型69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/44150API
curl https://kongchang.com/api/v1/knowledge/claims/44150MCP
get_claim(id=44150)