Embeddings with Meridian

Convert text into dense vector representations. One endpoint, sub‑10ms latency, dimensions from 256 to 3072.

Python

import requests

resp = requests.post(
    "https://api.meridian.ai/v1/embeddings",
    headers={"Authorization": "Bearer $MERIDIAN_KEY"},
    json={
        "model": "meridian-embed-v3",
        "input": "The quick brown fox",
        "encoding_format": "float",
        "dimensions": 1536,
    },
)
print(resp.json()["data"][0]["embedding"])

cURL

curl https://api.meridian.ai/v1/embeddings \
  -H "Authorization: Bearer $MERIDIAN_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meridian-embed-v3",
    "input": "The quick brown fox",
    "encoding_format": "float",
    "dimensions": 1536
  }'

Response

{
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [0.0023, -0.0091, ...]
    }
  ],
  "model": "meridian-embed-v3",
  "usage": { "prompt_tokens": 4, "total_tokens": 4 }
}