Expired50% confidenceFactExact time
Kimi K2.7 Code has a total of 61 layers, including 1 dense layer.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Kimi K2.7 Code Deep Dive: The Cost-Performance Battle of a Trillion-Parameter Coding Model
bilibiliAI睡前简报6/13/2026
Related Claims
UnverifiedKimi K2.7 Code uses a MoE architecture with 384 experts, selecting 8 experts plus 1 shared expert per token, with actual computation approximately 2.3% of that of a dense model.67% similarUnverifiedOn the Coder Reina coding leaderboard, Kimi K2.6 ranks near the top.66% similarUnverifiedK2.7-Code's total quality score is 153.5 points, while K2.6's total quality score is 79 points, with K2.7-Code leading by 74.5 points.64% similarVerifiedKimi K2.7 Code's chain-of-thought token usage is reduced by approximately 30% compared to K2.6.63% similarVerifiedKimi K2.6 adopts a Mixture of Experts (MoE) architecture, with a total parameter count at the trillion level and only 32B parameters actually activated.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/41730API
curl https://kongchang.com/api/v1/knowledge/claims/41730MCP
get_claim(id=41730)