Unverified50% confidenceFactExact time
The 1017 checkpoint's model architecture (552B MoE, 8B/16B activated), context window (1M), maximum output (384K), multimodal input and thinking enabled by default are exactly the same as the September version.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
Partially VerifiedStep-5-Preview adopts an MoE architecture with 600B total parameters and 27B activated parameters.70% similarUnverified千问3.8 27B基于千问3.5的整套架构体系，架构层面与3.5完全一致，仅做了继续预训练和后训练（含SFT与RL强化）68% 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.66% similarUnverifiedOllama Hub hosts a variety of models ranging from 7B to 70B parameters.66% similarVerifiedHy3 adopts a Mixture of Experts (MoE) architecture with a total parameter count of 295 billion (295B), activating only 21 billion (21B) parameters per inference.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067667API
curl https://kongchang.com/api/v1/knowledge/claims/1067667MCP
get_claim(id=1067667)