Fix langfuse prompt management on proxy (#7535)

* fix(types/utils.py): support langfuse + humanloop routes on llm router

* fix(main.py): remove acompletion elif block

just await if coroutine returned

litellm/main.py

@@ -435,51 +435,6 @@ async def acompletion(
         ctx = contextvars.copy_context()
         func_with_context = partial(ctx.run, func)
 
-        if (
-            custom_llm_provider == "openai"
-            or custom_llm_provider == "azure"
-            or custom_llm_provider == "azure_text"
-            or custom_llm_provider == "custom_openai"
-            or custom_llm_provider == "anyscale"
-            or custom_llm_provider == "mistral"
-            or custom_llm_provider == "openrouter"
-            or custom_llm_provider == "deepinfra"
-            or custom_llm_provider == "perplexity"
-            or custom_llm_provider == "groq"
-            or custom_llm_provider == "nvidia_nim"
-            or custom_llm_provider == "cohere_chat"
-            or custom_llm_provider == "cohere"
-            or custom_llm_provider == "cerebras"
-            or custom_llm_provider == "sambanova"
-            or custom_llm_provider == "ai21_chat"
-            or custom_llm_provider == "ai21"
-            or custom_llm_provider == "volcengine"
-            or custom_llm_provider == "codestral"
-            or custom_llm_provider == "text-completion-codestral"
-            or custom_llm_provider == "deepseek"
-            or custom_llm_provider == "text-completion-openai"
-            or custom_llm_provider == "huggingface"
-            or custom_llm_provider == "ollama"
-            or custom_llm_provider == "ollama_chat"
-            or custom_llm_provider == "replicate"
-            or custom_llm_provider == "vertex_ai"
-            or custom_llm_provider == "vertex_ai_beta"
-            or custom_llm_provider == "gemini"
-            or custom_llm_provider == "sagemaker"
-            or custom_llm_provider == "sagemaker_chat"
-            or custom_llm_provider == "anthropic"
-            or custom_llm_provider == "anthropic_text"
-            or custom_llm_provider == "predibase"
-            or custom_llm_provider == "bedrock"
-            or custom_llm_provider == "databricks"
-            or custom_llm_provider == "triton"
-            or custom_llm_provider == "clarifai"
-            or custom_llm_provider == "watsonx"
-            or custom_llm_provider == "cloudflare"
-            or custom_llm_provider == "aiohttp_openai"
-            or custom_llm_provider in litellm.openai_compatible_providers
-            or custom_llm_provider in litellm._custom_providers
-        ):  # currently implemented aiohttp calls for just azure, openai, hf, ollama, vertex ai soon all.
         init_response = await loop.run_in_executor(None, func_with_context)
         if isinstance(init_response, dict) or isinstance(
             init_response, ModelResponse
@@ -500,9 +455,6 @@ async def acompletion(
                 response_object=response,
                 model_response_object=litellm.ModelResponse(),
             )
-        else:
-            # Call the synchronous function using run_in_executor
-            response = await loop.run_in_executor(None, func_with_context)  # type: ignore
         if isinstance(response, CustomStreamWrapper):
             response.set_logging_event_loop(
                 loop=loop

litellm/proxy/_new_secret_config.yaml

@@ -3,6 +3,13 @@ model_list:
     litellm_params:
       model: openai/gpt-3.5-turbo
       api_key: os.environ/OPENAI_API_KEY
+  - model_name: chatbot_actions
+    litellm_params:
+      model: langfuse/azure/gpt-4o
+      api_base: "os.environ/AZURE_API_BASE"
+      api_key: "os.environ/AZURE_API_KEY"
+      tpm: 1000000
+      prompt_id: "jokes"
 
 litellm_settings:
   default_team_settings:

litellm/types/utils.py

@@ -1800,6 +1800,8 @@ class LlmProviders(str, Enum):
     INFINITY = "infinity"
     DEEPGRAM = "deepgram"
     AIOHTTP_OPENAI = "aiohttp_openai"
+    LANGFUSE = "langfuse"
+    HUMANLOOP = "humanloop"
 
 
 class LiteLLMLoggingBaseClass: