Unverified50% confidenceFactExact time
SigLIP2 SO400M采用ViT-SO400M架构，参数量约4亿，输出嵌入维度为1152
1
Sources
50%
Confidence
Long-term
Relevance
7/10/2026
First Seen
Sources
DINOv2 vs SigLIP: Why the Huge Gap in k-NN Classification? A Guide to Avoiding Vision Encoder Selection Pitfalls
redditr/MachineLearning7/8/2026
Related Claims
UnverifiedDINOv2 Giant参数量约11亿、嵌入维度1536，均显著大于SigLIP2 SO400M（约4亿参数、维度1152）和CLIP ViT-L（约3亿参数、维度768），但k-NN任务表现最差69% similarUnverifiedConsumer-grade GPUs such as the RTX 4090 have 24 GB of VRAM and can run quantized models with billions of parameters.67% similarVerifiedThe ESP32 is equipped with a dual-core Xtensa LX6 processor, a clock speed of up to 240 MHz, and 520 KB of SRAM.64% similarUnverified英伟达H100 GPU单卡功耗高达700瓦，一个配备5万块H100的超算集群总功耗可达35兆瓦（MW）64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/469426API
curl https://kongchang.com/api/v1/knowledge/claims/469426MCP
get_claim(id=469426)