Great Work Labs
Models / Search and retrieval

Atlas

Text embeddings
atlas-1

384-dimension embeddings in 100 languages, OpenAI-compatible.

Price$0.005 per 1,000 texts
Speed (our servers)22 to 46 ms
Languages~100 languages
InputText, batches up to 128
Built onintfloat/multilingual-e5-small (MIT)

Atlas turns text into 384-number vectors for semantic search, clustering, deduplication and recommendations. Similar meanings land close together across languages, so an English query finds a German document.

Set kind: "query" for search queries and leave the default passage for the documents you index. It also answers OpenAI-style requests at /v1/embeddings.

Use it

curl https://labs.greatwork.company/v1/run \
  -H "Authorization: Bearer $GWL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "atlas-1",
  "input": [
    "How do I reset my password?",
    "Où puis-je télécharger ma facture ?",
    "Shipping takes 3 to 5 business days."
  ]
}'

Set GWL_API_KEY to a key from your dashboard. The bare name atlas works too and follows new versions. Also answers POST /v1/embeddings (OpenAI): see compatible endpoints.

Options

input: A string, or an array of up to 128 strings. Texts over about 2,000 characters are truncated: split long documents first.

NameTypeDefaultWhat it does
kindstring
passage | query
"passage"query for search queries, passage for the documents you search over.

Output

FieldTypeMeaning
embeddingnumber[]384 floats, L2-normalized (dot product = cosine similarity).
dimsinteger384.

Every response wraps this in {"id", "model", "output", "usage": {"units", "unit", "cost_usd"}, "latency_ms"}. Send an array as input and output is an array in the same order.

Examples

Index documents

{
  "model": "atlas-1",
  "input": [
    "How do I reset my password?",
    "Où puis-je télécharger ma facture ?",
    "Shipping takes 3 to 5 business days."
  ]
}
[
  {
    "embedding": [
      0.074629,
      0.000765,
      -0.006023,
      -0.084802,
      0.053536,
      -0.06474,
      "... 378 more"
    ],
    "dims": 384
  },
  {
    "embedding": [
      0.038047,
      -0.003843,
      -0.047707,
      -0.092392,
      0.030254,
      -0.007609,
      "... 378 more"
    ],
    "dims": 384
  },
  {
    "embedding": [
      -0.005012,
      -0.005834,
      -0.02256,
      -0.060305,
      0.045396,
      -0.036372,
      "... 378 more"
    ],
    "dims": 384
  }
]

Captured on our production servers: 46 ms, 3 texts, $0.000015.

Search query

{
  "model": "atlas-1",
  "input": "forgot my login",
  "kind": "query"
}
{
  "embedding": [
    0.029984,
    0.001574,
    0.033558,
    -0.072531,
    0.046793,
    -0.034219,
    "... 378 more"
  ],
  "dims": 384
}

Captured on our production servers: 22 ms, 1 text, $0.000005.

Good for

  • Semantic search over docs and tickets
  • Find duplicate questions
  • Cluster feedback into themes
  • Recommend similar articles

Limits and caveats

  • Reads about the first 512 tokens: chunk long documents.
  • For final ranking quality, rerank the top results with Ladder.

Pricing

$0.005 per 1,000 texts. One unit per text up to 2,000 characters; longer texts are truncated, and billed one unit per started 2,000 characters. Failed calls aren't charged, and a call never costs more than its estimate (POST /v1/estimate tells you before you run it).

Built on

intfloat/multilingual-e5-small, 118M parameters. License: MIT. We run it on our servers.