Unverified50% confidenceFactExact time
MiniDog's transformer backbone has only 22 million parameters, far smaller than the billion-scale parameter counts typical of mainstream text-to-image models.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe smallest version of MiniMind has approximately 26 million to 64 million parameters.71% similarUnverifiedTiny-MoE模型总参数量约200M，但每个token激活的参数仅约90M67% similarUnverified轻量版模型单次推理的算力消耗往往只有旗舰版本的十分之一甚至更低66% similarUnverifiedThe TinyLoRA method had previously been shown to require fewer than 10 parameters for LLM fine-tuning to bring accuracy improvements.65% similarUnverifiedFor video generation models with billions of parameters, LoRA fine-tuning typically requires adjusting less than 1% of the parameters.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1026751API
curl https://kongchang.com/api/v1/knowledge/claims/1026751MCP
get_claim(id=1026751)