Unverified50% confidenceFactExact time
MiniLM is a lightweight Transformer architecture compressed from large models such as BERT via knowledge distillation, where L-6 denotes 6 layers and v2 is the second version.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMiniMax adopts a Transformer architecture combined with a Diffusion Model.73% similarUnverifiedMiniCPM-V 2.6 uses a dense Transformer architecture, achieving image-text understanding performance with approximately 8B parameters.69% similarUnverifiedMiniCPM-V is a lightweight model family for multimodal tasks, with previous versions known for achieving image-text understanding capabilities on edge devices.65% similarUnverifiedMiniCPM5 2B uses the standard Llama architecture, can be natively loaded by vLLM, and supports Think/No-Think dual inference modes and native tool calling.65% similarUnverifiedThe model used in this workflow is described as minimax ref2va pruned int8 convrot with turbo 4 step lora.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1028163API
curl https://kongchang.com/api/v1/knowledge/claims/1028163MCP
get_claim(id=1028163)