harrier-oss-v1-0.6b

Name: harrier-oss-v1-0.6b
Author: microsoft

by microsoft

Compact 600M multilingual text embedding with SOTA quality

203Kdl/month

600Mparams

HuggingFace Run it on your own data, free

Identifiers

Model ID

microsoft/harrier-oss-v1-0.6b

Feature URI

mixpeek://text_extractor@v1/microsoft_harrier_oss_v1_06b

Overview

Harrier OSS is a 600M-parameter text embedding model from Microsoft that achieves state-of-the-art performance on Multilingual MTEB v2 at release. Using decoder-only architecture with last-token pooling and knowledge distillation from larger models, it delivers embedding quality comparable to 8B-parameter models at a fraction of the compute cost.

With 30+ language support and MIT license, Harrier is ideal for production deployments where inference cost and latency matter -- 10x smaller than NVIDIA's Nemotron-8B with competitive retrieval quality.

Architecture

Decoder-only transformer with last-token pooling. 600M parameters. Knowledge-distilled from larger teacher models. Produces dense embeddings for text retrieval, classification, and clustering.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so harrier-oss-v1-0.6b 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 vector name has to match a vector index on the collection.
          vectors: { "text-embedding": yourVector },
          payload: { source_key: "archive/2026/asset-00412" },
        },
      ],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// text_extractor@v1 runs intfloat/multilingual-e5-large-instruct
// (1024-d) over a bucket, with no inference of your own.