All activity
Kimi K2.5 is Kimi's most intelligent model to date, achieving open-source SoTA performance in Agent, code, visual understanding, and a range of general intelligent tasks. It is also Kimi's most versatile model to date, featuring a native multimodal architecture that supports both visual and text input, thinking and non-thinking modes, and dialogue and Agent tasks.
Kimi K2.5Native multimodal model with self-directed agent swarms
Introducing Kimi Slides - Now with Nano Banana Pro š
šø Agentic search (Kimi K2)
šø Files ā Slides (PDFs, images, docs+)
šø Designer-level visuals (infographics, illustrations)
šø Fully editable + PPTX export
Kimi Slides - Now with Nano Banana ProFully Editable & Designer-Level Slides
š¹ SOTA on HLE (44.9%) and BrowseComp (60.2%)
š¹ Executes up to 200 ā 300 sequential tool calls without human interference
š¹ Excels in reasoning, agentic search, and coding
š¹ 256K context window
Kimi K2 ThinkingThe 1T Parameters Open-Source Thinking Model - SOTA on HLE