Merge branch 'main' into use-openai-for-databricks

tests/integration/inference/test_openai_embeddings.py

@@ -33,6 +33,7 @@ def skip_if_model_doesnt_support_user_param(client, model_id):
     provider = provider_from_model(client, model_id)
     if provider.provider_type in (
         "remote::together",  # service returns 400
+        "remote::fireworks",  # service returns 400 malformed input
     ):
         pytest.skip(f"Model {model_id} hosted by {provider.provider_type} does not support user param.")
 
@@ -42,6 +43,7 @@ def skip_if_model_doesnt_support_encoding_format_base64(client, model_id):
     if provider.provider_type in (
         "remote::together",  # param silently ignored, always returns floats
         "remote::databricks",  # param silently ignored, always returns floats
+        "remote::fireworks",  # param silently ignored, always returns list of floats
     ):
         pytest.skip(f"Model {model_id} hosted by {provider.provider_type} does not support encoding_format='base64'.")
 
@@ -289,7 +291,6 @@ def test_openai_embeddings_base64_batch_processing(compat_client, client_with_mo
         input=input_texts,
         encoding_format="base64",
     )
-
     # Validate response structure
     assert response.object == "list"
     assert response.model == embedding_model_id

tests/integration/suites.py

@@ -116,6 +116,15 @@ SETUP_DEFINITIONS: dict[str, Setup] = {
             "embedding_model": "databricks/databricks-bge-large-en",
         },
     ),
+    "fireworks": Setup(
+        name="fireworks",
+        description="Fireworks provider with a text model",
+        defaults={
+            "text_model": "accounts/fireworks/models/llama-v3p1-8b-instruct",
+            "vision_model": "accounts/fireworks/models/llama-v3p2-90b-vision-instruct",
+            "embedding_model": "nomic-ai/nomic-embed-text-v1.5",
+        },
+    ),
 }