Moonshot AI releases open-source multimodal agent model Kimi K2.5
Original titlemoonshotai/Kimi-K2.5
Moonshot AI released Kimi K2.5, an open-source native multimodal agentic model built by continual pretraining on about 15 trillion mixed visual and text tokens.
The model card reports a 1T-parameter Mixture-of-Experts architecture with 32B activated parameters and a 256K context length, and it lists benchmark results against GPT-5.2, Claude 4.5 Opus, Gemini 3 Pro, DeepSeek V3.2, and Qwen3-VL-235B-A22B-Thinking.
Weights and code are released under a Modified MIT License, with API access on the Moonshot platform.
The model card gives a full benchmark table against GPT-5.2, Claude 4.5 Opus, and Gemini 3 Pro, useful for comparing open multimodal agent models.
Source: Moonshot AI (Kimi) · new models on Hugging Face · huggingface.coPublished · added here