Unverified50% confidenceFactExact time
According to the documentation, Kimi K2.8 performs at a level comparable to K3 but has more efficient reasoning capabilities, supporting a 1-million-token context window, image and video input, and three reasoning modes: low, high, and max.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/16/2026
First Seen
Valid until: 12/15/2026
Sources
Related Entities
Related Claims
UnverifiedKimi K3使用自研的Kimi Delta Attention机制，解码速度最高提升6.3倍，上下文窗口支持100万token78% similarUnverifiedKimi K3 was released on July 16 and integrated into Kimi Code, featuring a parameter scale of 2.8 trillion, supporting up to 1 million token context and native visual understanding.73% similarUnverifiedKimi K2 demonstrated faster execution speed during testing, quickly showing a Complete status.70% similarUnverifiedKimi K3's 2.8 trillion parameter scale surpasses the previously largest open-weight model, DeepSeek-V3, which had 671 billion parameters.70% similarUnverifiedAccording to official data, K3 achieves approximately a 2.5× improvement in overall inference efficiency compared to K2.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/927488API
curl https://kongchang.com/api/v1/knowledge/claims/927488MCP
get_claim(id=927488)