phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 03:04:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat(databricks/chat): support structured outputs on databricks

Browse source 
Closes https://github.com/BerriAI/litellm/pull/6978

- handles content as list for dbrx, - handles streaming+response_format for dbrx
This commit is contained in:
Krrish Dholakia 2024-12-02 18:23:05 -08:00
parent 12aea45447
commit 0caf804f4c
18 changed files with 538 additions and 193 deletions
Download patch file Download diff file Expand all files Collapse all files

31
tests/local_testing/test_embedding.py
View file

@ -932,37 +932,6 @@ async def test_gemini_embeddings(sync_mode, input):
pytest.fail(f"Error occurred: {e}")
@pytest.mark.parametrize("sync_mode", [True, False])
@pytest.mark.asyncio
async def test_databricks_embeddings(sync_mode):
try:
litellm.set_verbose = True
litellm.drop_params = True
if sync_mode:
response = litellm.embedding(
model="databricks/databricks-bge-large-en",
input=["good morning from litellm"],
instruction="Represent this sentence for searching relevant passages:",
)
else:
response = await litellm.aembedding(
model="databricks/databricks-bge-large-en",
input=["good morning from litellm"],
instruction="Represent this sentence for searching relevant passages:",
)
print(f"response: {response}")
openai.types.CreateEmbeddingResponse.model_validate(
response.model_dump(), strict=True
)
# stubbed endpoint is setup to return this
# assert response.data[0]["embedding"] == [0.1, 0.2, 0.3]
except Exception as e:
pytest.fail(f"Error occurred: {e}")
# test_voyage_embeddings()
# def test_xinference_embeddings():
# try: