Moonshot AI's Kimi K3 Model Surpasses Claude Fable 5 in Benchmark
Moonshot AI introduces Kimi K3, a 2.8 trillion parameter model excelling in coding benchmarks.
Moonshot AI has unveiled the Kimi K3 model, boasting 2.8 trillion parameters. K3 outperformed several models, including Claude Fable 5, in coding and agentic benchmarks. The model features a 1 million token context window and activates only 16 of its 896 experts per token. With a score of 1679, K3 ranked first in the Frontend Code Arena.