How could I create custom embedder?

[stimw]

I've read through the documentation but didn’t find any info about how to create a custom embedder.

For example, I'm using SiliconFlow’s embedding API:

curl --request POST \
  --url https://api.siliconflow.cn/v1/embeddings \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "BAAI/bge-large-zh-v1.5",
  "input": "Silicon flow embedding online: fast, affordable, and high-quality embedding services. come try it out!"
}'

Response:

{
  "model": "<string>",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        123
      ],
      "index": 123
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  }
}

[yuvarajsl]

Hey @stimw,
Thanks for reaching out and supporting Agno. I’ve shared this with the team, we’re working through all requests one by one and will get back to you soon.
If it’s urgent, please let us know. We appreciate your patience!

[Mustafa-Esoofally]

Hey @stimw ! The siliconflow models are OpenAI-compatible. Here's how you can use them:

from agno.knowledge.embedder.openai import OpenAIEmbedder

embedder = OpenAIEmbedder(
    id="<Embedding model>,
    dimensions=1024,  # dimensions might need to be updated
    api_key="<API_KEY>",
    base_url="https://api.siliconflow.com/v1", 
)

text = "Silicon flow embedding online: fast, affordable, and high-quality embedding services. come try it out!"

# Test get_embedding
embedding = embedder.get_embedding(text)