I’m following the docs to produce embeddings in a pro-code way (as an alternative to use Pipeline builder, which supports creating embeddings with in a no-code transform).
When I follow the docs, the limit that are present in the examples are:
RATE_LIMIT_PER_MIN = 100
TOKEN_LIMIT_PER_MIN = 50000
Can those limit be raised ? Be changed ? If I were to reimplement the logic without using an orchestrator, do I need to comply with those limits ?