Kimi K2.6

Kimi K2.6 is a multimodal model from Moonshot AI, positioned as a workhorse for agent loops, coding tasks, and visual context. It is the first version in the kimi-k2 line. It has a context window of 262144 tokens and a maximum output of 262144 tokens. Weights are open. It supports reasoning mode and tool calling. It accepts image, text, and video inputs and returns text. Pricing starts at $0.275 per million input tokens and $1.10 per million output tokens, with cached input at $0.05 per million tokens.

First version of its line in the registry.

Range across the hosts that serve it: up to $1.79 per 1M input tokens.

• ·262144-token context and output window
• ·Open weights, reasoning mode, and tool calling
• ·Accepts image, text, and video inputs
• ·Served by 59 hosts for availability
• ·Knowledge cutoff January 2025

• Reach for it for agent loops and coding tasks
• Reach for it for visual context in multimodal inputs
• Reach for it for long-context reasoning with 262k tokens
• Reach for it for tool-calling workflows

59 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window and output limit?

Both context and max output are 262144 tokens.

Is Kimi K2.6 open weights?

Yes, the weights are open.

What input modalities does it support?

It accepts image, text, and video, and returns text.