Kimi-K2.6

Name: Kimi-K2.6
Author: moonshotai

by moonshotai

1T-parameter MoE multimodal model with 32B active parameters

2.7Mdl/month

1T total (32B active, MoE)params

HuggingFace Run it on your own data, free

Identifiers

Model ID

moonshotai/Kimi-K2.6

Feature URI

mixpeek://image_extractor@v1/moonshotai_kimi_k26_v1

Overview

Kimi-K2.6 is a massive Mixture-of-Experts model from Moonshot AI with 1 trillion total parameters and 32 billion active parameters per forward pass. It features native multimodal capabilities via a MoonViT 400M vision encoder, achieving state-of-the-art results on mathematical reasoning (MathVision 93.2%) and multimodal understanding (MMMU-Pro 79.4%). Its MIT-like license makes it one of the most capable openly-licensed models available.

Architecture

Sparse Mixture-of-Experts architecture with 1T total parameters, 32B active per token. Uses MoonViT-400M as the vision encoder for native image understanding. Supports 256K context length. MoE routing enables efficient inference despite massive parameter count.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so Kimi-K2.6 runs
// on your side and the output is upserted through POST
// /v1/namespaces/{namespace_id}/documents/upsert. On Enterprise the other
// path is to upload the weights instead: POST /v1/namespaces/{id}/models
// accepts the huggingface format and a custom plugin loads them.
const res = await fetch(
  "https://api.mixpeek.com/v1/namespaces/ns_your_namespace/documents/upsert",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      collection_id: "col_your_collection",
      documents: [
        {
          document_id: "asset-00412",
          // The model produces text, so it lands in payload. Give the
          // collection a text vector index and embed that text to make it
          // searchable rather than only filterable.
          payload: { extracted_text: modelOutput, source_key: "archive/2026/asset-00412" },
          vectors: { "multimodal-embedding": embeddingOfModelOutput },
        },
      ],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// universal_extractor@v1 runs google/gemini-embedding-2
// (3072-d) over a bucket, with no inference of your own.