Kimi K2.6

Capabilities

Context Window

262,144 tokens

Max Output

262,144 tokens

Knowledge Cutoff

April 2025

Modalities

Input: Text

Output: Text

Latency

1.8 s

Median time to first token over the last 30 days (352 samples).

Pricing Details

Technical Specs

Total Parameters

1T

Activated Parameters

32B

Layers

61

Dense Layers

1

Attention Hidden Dimension

7168

MoE Hidden Dimension (per Expert)

2048

Attention Heads

64

Experts

384

Selected Experts per Token

8

Shared Experts

1

Vocabulary Size

160K

Attention Mechanism

MLA

Activation Function

SwiGLU

Vision Encoder

MoonViT

Vision Encoder Parameters

400M