Moonshot's Kimi K2.5 open model beats Sonnet at half the cost
Moonshot AI releases Kimi K2.5, a 32B active-parameter open-weights multimodal model claiming benchmark results above Claude Sonnet at lower API cost.
MoonshotAI's Kimi K2.5 is a 32B active-1T parameter open-weights model featuring native multimodality with image and video understanding, built through continual pretraining on 15 trillion mixed visual and text tokens. It introduces a new MoonViT vision encoder and supports advanced capabilities like Agent Swarm, which coordinates up to 100 sub-agents for parallel workflows, and an Office Productivity K2.5 Agent for large-scale office tasks.
This release marks a significant leap in open models from China, claiming state-of-the-art results on benchmarks like HLE and BrowseComp, and offering aggressive API pricing and throughput.