0.1.3

EmpirioLabsEmbeddings uses LangChain's default chunk_size of 1,000 texts again: the EmpirioLabs AI API now takes up to 2,048 inputs per request for every embedding model, so embed_documents makes fewer, larger requests than in 0.1.2.

0.1.2

EmpirioLabsEmbeddings sends 10 texts per request by default, the most text-embedding-v4 accepts, so embed_documents works for any number of texts. Set chunk_size for a model that takes more. The docstrings describe EmpirioLabs AI with the same wording as empiriolabs.ai.

0.1.1

Describes EmpirioLabs AI with the same wording as empiriolabs.ai, lists the current DeepSeek V4.1 Flash model id, and links the EmpirioLabs-AI repository.