Unverified50% confidenceFactExact time
Spark X 2.5 has 36 layers in total and uses a repeating '4-layers-per-group' structure, where each group contains three sliding window layers that attend only to the most recent 512 tokens and one global layer that reads the entire text
1
Sources
50%
Confidence
Long-term
Relevance
9/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAgnes-3.0-Flash has a total of 72 layers, consisting of 54 delta-rule recurrent layers and 18 global attention layers arranged in a 3:1 alternating pattern, with a hidden dimension of 5120 and a vocabulary size of 248,32056% similarUnverifiedMuse Spark 1.3 scored 57/80 (71.25%) on KingBench 3, a regression from version 1.2's 76.25%.52% similarUnverifiedOn the 3D watch problem, the previous all-time high score was 8 points, set by Muse Spark 1.3.51% similarUnverifiedThe model ultimately outputs a tensor of 201×18×4, where 4 represents 4 lane lines, 18 represents key rows, and 201 represents 200 grid positions plus 1 non-existence class.51% 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.51% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/954246API
curl https://kongchang.com/api/v1/knowledge/claims/954246MCP
get_claim(id=954246)