fix: make vision and embedding tests pass with openai, anthropic and gemini

NOTE - Anthropic embeddings do not work due to LiteLLM not supporting them.
2025-02-26 10:52:33 -08:00 · 2025-02-26 10:52:33 -08:00 · 4cf95475e5
commit 4cf95475e5
parent abfc4b3bce
4 changed files with 10 additions and 5 deletions
--- a/llama_stack/providers/utils/inference/litellm_openai_mixin.py
+++ b/llama_stack/providers/utils/inference/litellm_openai_mixin.py
@ -97,7 +97,6 @@ class LiteLLMOpenAIMixin(
        )

        params = await self._get_params(request)
-
        # unfortunately, we need to use synchronous litellm.completion here because litellm
        # caches various httpx.client objects in a non-eventloop aware manner
        response = litellm.completion(**params)