What's the difference between ollama.embeddings() and ollama.embed() ? Why do the methods return different embeddings for the same model (code in description)?

[rustam-007]

Hello! I am calling both methods to compare the embeddings they return.

ll = ollama.embeddings(model='llama3.2',
prompt = 'The sky is blue because of rayleigh scattering'
)
llm = dict(ll)
llm['embedding']

ll = ollama.embed(model='llama3.2',
input = 'The sky is blue because of rayleigh scattering'
)
llm = dict(ll)
llm['embeddings'][0]

They return different embeddings for the same model. Why is that?

Thank you in advance

[hopperelec]

embeddings is a deprecated implementation, I believe only left for backwards compatibility. You should usually just use embed.