Unverified50% confidenceFactExact time
The smallest version of MiniMind has approximately 26 million to 64 million parameters.
1
Sources
50%
Confidence
Long-term
Relevance
9/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTiny-MoE模型总参数量约200M，但每个token激活的参数仅约90M69% similarUnverifiedA model with 11,000 parameters occupies less than 50KB of memory.68% similarUnverifiedWhisper's tiny or base variants range from 39M to 74M parameters, and combined with INT8 quantization, the model size can be compressed to 1/4 of the original.67% similarUnverifiedA model with 100 billion parameters requires approximately 80 GB of VRAM to process a single inference request, while a model with tens of billions of parameters requires only about 20 GB.65% similarUnverified谷歌DeepMind团队在2022年的研究中记录了超过150种在小模型中缺失、却在大模型中出现的能力，包括多步推理、类比迁移与少样本学习65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/891491API
curl https://kongchang.com/api/v1/knowledge/claims/891491MCP
get_claim(id=891491)