By capability
Embeddings
No model serves this yet
Nothing in the model catalogue supports this route right now. The page stays for reference.
#Embed text
Same shape as OpenAI: input takes a string or an array of strings, the response carries one vector per input in the same order.
embeddings.sh
curl https://www.aisplice.store/v1/embeddings \
-H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{
"model": "text-embedding-3-small",
"input": ["hello world", "another sentence"]
}'#Batching and dimensions
Batch multiple inputs in one call to save round trips — embeddings bill by input tokens, so batching costs the same but is much faster. Models that support Matryoshka truncation accept a dimensions parameter to shrink the vector.
Still stuck?
Send us the request ID or the error message and we will look into it.