some fixes

autollm/auto/llm.py

@@ -12,9 +12,9 @@ class AutoLiteLLM:
     @staticmethod
     def from_defaults(
             model: str = "gpt-3.5-turbo",
-            api_base: Optional[str] = None,
-            max_tokens: Optional[int] = None,
-            temperature: float = 0.1) -> LLM:
+            max_tokens: Optional[int] = 256,
+            temperature: float = 0.1,
+            api_base: Optional[str] = None) -> LLM:
         """
         Create any LLM by model name. Check https://docs.litellm.ai/docs/providers for a list of
         supported models.
@@ -25,13 +25,12 @@ def from_defaults(
         Parameters:
             model: Name of the LLM model to be initialized. Check
         https://docs.litellm.ai/docs/providers for a list of supported models.
-            api_base: The API base URL to use for the LLM.
             max_tokens: The maximum number of tokens to generate by the LLM.
             temperature: The temperature to use when sampling from the distribution.
-            **kwargs: Arbitrary keyword arguments.
+            api_base: The API base URL to use for the LLM.
 
         Returns:
             LLM: The initialized LiteLLM instance for given model name and parameter set.
         """
 
-        return LiteLLM(model=model, api_base=api_base, max_tokens=max_tokens, temperature=temperature)
+        return LiteLLM(model=model, max_tokens=max_tokens, temperature=temperature, api_base=api_base)

autollm/auto/query_engine.py

@@ -14,9 +14,9 @@ def create_query_engine(
         documents: Optional[Sequence[Document]] = None,
         # llm_params
         llm_model: str = "gpt-3.5-turbo",
-        llm_api_base: Optional[str] = None,
-        llm_max_tokens: Optional[int] = None,
+        llm_max_tokens: Optional[int] = 256,
         llm_temperature: float = 0.1,
+        llm_api_base: Optional[str] = None,
         # service_context_params
         system_prompt: str = None,
         query_wrapper_prompt: str = None,

tests/config.yaml

@@ -1,24 +1,17 @@
 # config.example.yaml
 tasks:
   - name: "summarize"
+    llm_model: "gpt-3.5-turbo"
+    llm_max_tokens: 1028
+    llm_temperature: 0.1
     system_prompt: "You are a friendly chatbot that can summarize documents.:"  # System prompt for this task
-    vector_store_params:
-      vector_store_type: "SimpleVectorStore"
-    llm_params:
-      model: "gpt-3.5-turbo"
-    service_context_params:
-      chunk_size: 1024
-    query_engine_params:
-      similarity_top_k: 5
     enable_cost_calculator: true
+    embed_model: "default"
+    chunk_size: 512
+    chunk_overlap: 64
+    context_window: 2048
+    similarity_top_k: 2
+    vector_store_type: "SimpleVectorStore"
   - name: "qa"
     system_prompt: "You are a friendly chatbot that can answer questions."  # System prompt for this task
-    vector_store_params:
-      vector_store_type: "SimpleVectorStore"
-    llm_params:
-      model: "gpt-3.5-turbo"
-    service_context_params:
-      chunk_size: 1024
-    query_engine_params:
-      similarity_top_k: 3
     enable_cost_calculator: false