What endpoint to use for embeddings?

[jamesbraza]

I am using the bert-embeddings model per here.

https://localai.io/features/embeddings/ talks about the bert-embeddings backend, but doesn't talk about the API spec anywhere or have a curl example. Looking at the source here, I can see there are several, but I am not sure:

• Which one (/embeddings vs /v1/embeddings) to use
• What JSON blob structure to send to them

What endpoint should be used for embeddings? Can you provide me a curl example command?

[jamesbraza]

Okay, I figured this out:

> curl http://localhost:8080/embeddings -X POST -H "Content-Type: application/json" -d '{
    "input": "Your text string goes here",
    "model": "bert-embeddings"
  }' | jq "."
...
{
  "object": "list",
  "model": "bert-embeddings",
  "data": [
    {
      "embedding": [
        0.051830754,
        ...,
        -0.023937061
      ],
      "index": 0,
      "object": "embedding"
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0
  }
}

What do you think of me adding this to the docs?