Embeddings
Turn text into vectors for RAG, semantic search, classification, clustering. Same OpenAI shape — a drop-in for OpenAI-embedding migrators.
Endpoint
POSThttps://api.ollima.com/v1/embeddings
Worked example
resp = client.embeddings.create( model="nomic-embed-text", input=["DNS resolves names to IPs.", "BGP advertises reachability."], ) vectors = [d.embedding for d in resp.data] # vectors[0] is the 768-dim float list for the first input
Models
| Alias | Dimensions | Best for |
|---|---|---|
nomic-embed-text | 768 | General-purpose RAG. Cheap default. |
bge-large-en-v1.5 | 1024 | Higher quality, slightly more expensive. |
| OpenAI embedding via BYOK | 1536 | If you need OpenAI parity. |
Batched calls
Pass an array of strings in input — one request, one bill, n vectors back. Up to 2048 inputs per call.