{
  "$schema": "https://modeldeprecations.dev/api/v1/schema.json",
  "provider": "moonshot",
  "model": "kimi-k2.6",
  "name": "Kimi K2.6",
  "aliases": [],
  "description": "A 256k multimodal model taking text, image and video input, with thinking as a request-level mode rather than a separate id - the change that made kimi-k2-thinking and its turbo variant redundant. It is the general-purpose model in the current range, sitting below K3 and beside the code-specific K2.7s.",
  "status": "active",
  "replacements": [],
  "sources": [
    {
      "url": "https://platform.kimi.ai/docs/models",
      "title": "Kimi model list",
      "accessed": "2026-08-03"
    }
  ],
  "last_verified": "2026-08-03",
  "computed_status": "active",
  "asOf": "2026-08-03"
}