Moonshot AI

MoonshotAI: Kimi K3 is Kimi’s most powerful model to date, with 2.8 trillion parameters. It is built based on the hybrid linear attention mechanism Kimi Delta Attention and attention residual technology. It has native visual understanding capabilities and a context window of 1 million tokens. It is suitable for cutting-edge intelligent scenarios such as software engineering, knowledge work, and deep reasoning.

Input:

Output:

Input:

$3$0.9/1M

Output:

$15$4.5/1M

Context length:

1.05M

Max output:

1.05M

MoonshotAI: Kimi K3.1 Pro is MoonshotAI’s flagship inference model with a 1 million token context window. It is designed for long-range agent workflows and supports multi-modal understanding of image and video inputs. The model excels in programming, 3D authoring, and financial research, and supports tool invocation capabilities for streaming output, structured generation, and configurable inference capabilities.

Input:

Output:

Input:

—

Output:

—

Context length:

1.05M

Max output:

1.05M

MoonshotAI: Kimi K3.1 Pro Highspeed is a high-speed version of Kimi K3.1 Pro. While providing the same level of intelligence, the output speed is increased by about 2–3 times and supports context windows of up to 1M tokens. It is optimized for long-range reasoning and agent workflows, and excels in programming, 3D authoring, financial analysis, tool usage, and multimodal understanding of text, images, and video.

Input:

Output:

Input:

—

Output:

—

Context length:

1.05M

Max output:

1.05M