Unverified50% confidenceFactExact time
Krea 2 itself has 12B parameters, while this new model is only 7B.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/18/2026
First Seen
Valid until: 12/17/2026
Sources
Related Entities
Related Claims
UnverifiedThe 14B MoE model is a jailbroken version, with a KLD (divergence metric from the original model) of only 0.0015 (approximately 0.15%)68% similarUnverified7B-13B级别的开源小模型（如Llama 2 7B、Mistral 7B）参数容量有限，对外部提供的上下文质量形成刚性依赖63% similarUnverified万象模型提供了 1.3B 和 14B 参数量的不同规模版本63% similarVerifiedDeepSeek-V3 has 671B total parameters, but only approximately 37B parameters are activated per inference.60% 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.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/933004API
curl https://kongchang.com/api/v1/knowledge/claims/933004MCP
get_claim(id=933004)