Add /embeddings api

[prabirshrestha]

Add support for embedding api. https://platform.openai.com/docs/guides/embeddings/what-are-embeddings. This can be useful for implementing RAG and vector search.

Might be could start with nomic-embed-text-v1

[francis2tm]

we'll definitely add it. It seems llama.cpp is almost supporting embeddings.

[francis2tm]

@pedro-devv ggerganov/llama.cpp#5423
I guess we can get started

[prabirshrestha]

Nomic Embed also seems to be supported now by llama.cpp ggerganov/llama.cpp#5468

[AndriyMulyar]

You should use nomic-embed-text-v1.5 https://blog.nomic.ai/posts/nomic-embed-matryoshka

here is the v1.5 GGUF: https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF

[prabirshrestha]

ollama added support for nomic-embed-text as it is not supported by llama.cpp but unforutnately it is not compatible with OpenAI apis. Perf is really good too. https://ollama.com/library/nomic-embed-text

curl http://localhost:11434/api/embeddings -d '{
  "model": "nomic-embed-text",
  "prompt": "The sky is blue because of Rayleigh scattering"
}'