Researchers at Moonshot AI introduced Kimi Linear, a hybrid linear attention architecture that outperforms full attention across short-context, long-context, and reinforcement learning scaling regimes…
Moonshot's Kimi K3, a 2.8-trillion-parameter mixture-of-experts model released July 16, achieves a 1M-token context window with decoding up to 6.3x faster than its predecessor by using Kimi Delta Atte…
Moonshot AI released Kimi K3, a 2.8 trillion parameter open Mixture-of-Experts model with a 1 million token context window. The model introduces innovations like Kimi Delta Attention for faster decodi…
Chinese AI lab Moonshot AI launched the 2.8 trillion-parameter Kimi K3 open weights model on Thursday, with benchmark data showing it trails Anthropic and OpenAI's leading models by just one generatio…
Moonshot AI has unveiled Kimi K3, a flagship open-source AI model with 2.8 trillion parameters and a one-million-token context window, achieving frontier-level benchmark results. The full model weight…
Moonshot AI has launched Kimi K3, a 2.8-trillion-parameter mixture-of-experts model with a 1-million-token context window and native vision capabilities. The Beijing-based startup claims it challenges…
Researchers introduced Semidirect Fourier Delta Attention (SFDA), a phase-controlled generalization of Kimi Delta Attention that replaces real diagonal decay with block-rotational Fourier control, ach…
The Kimi Team published a new attention architecture, Kimi Linear, on arXiv in 2025. The architecture uses Kimi Delta Attention interleaved 3:1 with full attention, becoming the first linear hybrid to…
Researchers introduced Gated DeltaNet-2, a linear attention model that decouples the erase and write operations in recurrent state updates using separate channel-wise gates. The model outperforms Mamb…