Expired85% confidenceFactTime unknown
In the benchmark tests, each repository was run 4 times on each side and the median was taken.
1
Sources
85%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn mixed-precision training, each FP16 parameter occupies 2 bytes of GPU memory, and each FP32 parameter occupies 4 bytes of GPU memory.67% similarUnverified叠加cv=5的5折交叉验证后，144种参数组合需训练144×5=720次模型63% similarUnverifiedV4 Pro Max满血推理模式下算法竞赛3206分、编程93.5分、科学问答90.1分、软件工程80.6分、终端操作67.9分63% similarUnverifiedIn LOD generation testing, Mixar generated three levels—LOD0, LOD1, and LOD2—with textures reduced from 4K to 2K and 1K, creating a total of 52 new textures, with results meeting expectations.62% similarUnverifiedRankSpot received a rating of 4.5 on Product Hunt and has approximately 1,000 followers.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/17481API
curl https://kongchang.com/api/v1/knowledge/claims/17481MCP
get_claim(id=17481)