check model kwarg in tools

packages/jhehemann/customs/prediction_sum_url_content/prediction_sum_url_content.py

@@ -81,8 +81,8 @@ def count_tokens(text: str, model: str) -> int:
     "prediction-online-sum-url-content",
 ]
 TOOL_TO_ENGINE = {
-    "prediction-offline-sum-url-content": "gpt-4",
-    "prediction-online-sum-url-content": "gpt-4",
+    "prediction-offline-sum-url-content": "gpt-4-0125-preview",
+    "prediction-online-sum-url-content": "gpt-4-0125-preview",
 }
 
 
@@ -977,7 +977,7 @@ def fetch_additional_information(
     google_api_key: str,
     google_engine: str,
     nlp,
-    engine: str = "gpt-3.5-turbo",
+    engine: str = "gpt-4-0125-preview",
     temperature: float = 1.0,
     max_compl_tokens: int = 500,
 ) -> str:
@@ -1092,7 +1092,7 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         nlp = spacy.load("en_core_web_sm")
 
         # Get the LLM engine to be used
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
         print(f"ENGINE: {engine}")
 
         # Extract the event question from the prompt

packages/napthaai/customs/prediction_request_rag/prediction_request_rag.py

@@ -66,12 +66,12 @@ def __exit__(self, exc_type, exc_value, traceback) -> None:
 }
 MAX_TOKENS = {
     "gpt-3.5-turbo-0125": 16385,
-    "gpt-4": 8192,
+    "gpt-4-0125-preview": 8192,
 }
 ALLOWED_TOOLS = [
     "prediction-request-rag",
 ]
-TOOL_TO_ENGINE = {tool: "gpt-3.5-turbo-0125" for tool in ALLOWED_TOOLS}
+TOOL_TO_ENGINE = {tool: "gpt-4-0125-preview" for tool in ALLOWED_TOOLS}
 DEFAULT_NUM_URLS = defaultdict(lambda: 3)
 DEFAULT_NUM_QUERIES = defaultdict(lambda: 3)
 NUM_URLS_PER_QUERY = 5
@@ -596,7 +596,8 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
         additional_information, counter_callback = fetch_additional_information(
             client=client,
             prompt=prompt,

packages/napthaai/customs/prediction_request_reasoning/prediction_request_reasoning.py

@@ -691,8 +691,8 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         google_engine_id = api_keys.get("google_engine_id", None)
         temperature = kwargs.get("temperature", DEFAULT_OPENAI_SETTINGS["temperature"])
         max_tokens = kwargs.get("max_tokens", DEFAULT_OPENAI_SETTINGS["max_tokens"])
-        engine = TOOL_TO_ENGINE[tool]
-
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 

packages/napthaai/customs/prediction_url_cot/prediction_url_cot.py

@@ -41,12 +41,12 @@ def __exit__(self, exc_type, exc_value, traceback) -> None:
 }
 MAX_TOKENS = {
     "gpt-3.5-turbo-0125": 16385,
-    "gpt-4": 8192,
+    "gpt-4-0125-preview": 8192,
 }
 ALLOWED_TOOLS = [
     "prediction-url-cot",
 ]
-TOOL_TO_ENGINE = {tool: "gpt-3.5-turbo-0125" for tool in ALLOWED_TOOLS}
+TOOL_TO_ENGINE = {tool: "gpt-4-0125-preview" for tool in ALLOWED_TOOLS}
 DEFAULT_NUM_URLS = defaultdict(lambda: 3)
 DEFAULT_NUM_QUERIES = defaultdict(lambda: 3)
 NUM_URLS_PER_QUERY = 5
@@ -429,7 +429,7 @@ def fetch_additional_information(
 def adjust_doc_tokens(
     doc: Document, 
     max_tokens: int, 
-    engine: str = "gpt-3.5-turbo"
+    engine: str = "gpt-4-0125-preview"
 ) -> Document:
     """Adjust the number of tokens in the document."""
     if count_tokens(doc.text, engine) > max_tokens:
@@ -532,8 +532,8 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 
-        engine = TOOL_TO_ENGINE[tool]
-
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
         # fetch additional information from the web
         additional_information, counter_callback = fetch_additional_information(
             client=client,

packages/nickcom007/customs/prediction_request_sme/prediction_request_sme.py

@@ -68,16 +68,16 @@ def count_tokens(text: str, model: str) -> int:
     "temperature": 0.,
 }
 MAX_TOKENS = {
-    "gpt-3.5-turbo": 4096,
-    "gpt-4": 8192,
+    "gpt-3.5-turbo-0125": 4096,
+    "gpt-4-0125-preview": 8192,
 }
 ALLOWED_TOOLS = [
     "prediction-offline-sme",
     "prediction-online-sme",
 ]
 TOOL_TO_ENGINE = {
-    "prediction-offline-sme": "gpt-3.5-turbo",
-    "prediction-online-sme": "gpt-3.5-turbo",
+    "prediction-offline-sme": "gpt-4-0125-preview",
+    "prediction-online-sme": "gpt-4-0125-preview",
 }
 
 PREDICTION_PROMPT = """
@@ -435,7 +435,8 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
 
         try:
             sme, sme_introduction, counter_callback = get_sme_role(

packages/nickcom007/customs/sme_generation_request/sme_generation_request.py

@@ -62,8 +62,8 @@ def count_tokens(text: str, model: str) -> int:
 ]
 
 TOOL_TO_ENGINE = {
-    "strong-sme-generator": "gpt-4",
-    "normal-sme-generator": "gpt-3.5-turbo",
+    "strong-sme-generator": "gpt-4-0125-preview",
+    "normal-sme-generator": "gpt-3.5-turbo-0125",
 }
 
 SME_GENERATION_SYSTEM_PROMPT = """
@@ -125,7 +125,8 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"tool must be one of {ALLOWED_TOOLS}")
 
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
 
         market_question = SME_GENERATION_MARKET_PROMPT.format(question=prompt)
         system_prompt = SME_GENERATION_SYSTEM_PROMPT

packages/polywrap/customs/prediction_with_research_report/prediction_with_research_report.py

@@ -528,8 +528,8 @@ def run(**kwargs) -> Tuple[Optional[str], Any, Optional[Dict[str, Any]], Any]:
     if tool not in ALLOWED_TOOLS:
         raise ValueError(f"TOOL {tool} is not supported.")
 
-    model = TOOL_TO_ENGINE[tool]
-
+    model = kwargs.get("model", TOOL_TO_ENGINE[tool])
+    print(f"ENGINE: {model}")
     queries, counter_callback = generate_subqueries(query=prompt, limit=initial_subqueries_limit, api_key=openai_api_key, model=model, counter_callback=counter_callback)
     queries, counter_callback = rerank_subqueries(queries=queries, goal=prompt, api_key=openai_api_key, model=model, counter_callback=counter_callback)
     queries = queries[:subqueries_limit]

packages/psouranis/customs/optimization_by_prompting/optimization_by_prompting.py

@@ -83,8 +83,8 @@ def count_tokens(text: str, model: str) -> int:
     "deepmind-optimization",
 ]
 TOOL_TO_ENGINE = {
-    "deepmind-optimization-strong": "gpt-4",
-    "deepmind-optimization": "gpt-3.5-turbo",
+    "deepmind-optimization-strong": "gpt-4-0125-preview",
+    "deepmind-optimization": "gpt-3.5-turbo-0125",
 }
 
 PREDICTION_PROMPT_INSTRUCTIONS = """
@@ -222,7 +222,7 @@ def prompt_engineer(
     init_instructions,
     instructions_format,
     iterations=3,
-    model_name="gpt-3.5-turbo",
+    model_name="gpt-4-0125-preview",
 ):
 
     llm = OpenAILLM(model_name=model_name, openai_api_key=openai_api_key)
@@ -402,7 +402,8 @@ def run(**kwargs) -> Tuple[Optional[str], Optional[Dict[str, Any]], Any, Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
         additional_information = fetch_additional_information(
             prompt=prompt,
             engine=engine,

packages/valory/customs/prediction_request/prediction_request.py

@@ -81,10 +81,10 @@ def count_tokens(text: str, model: str) -> int:
     "prediction-online-summarized-info",
 ]
 MAX_TOKENS = {
-    "gpt-3.5-turbo": 4096,
-    "gpt-4": 8192,
+    "gpt-3.5-turbo-0125": 4096,
+    "gpt-4-0125-preview": 8192,
 }
-TOOL_TO_ENGINE = {tool: "gpt-3.5-turbo" for tool in ALLOWED_TOOLS}
+TOOL_TO_ENGINE = {tool: "gpt-4-0125-preview" for tool in ALLOWED_TOOLS}
 # the default number of URLs to fetch online information for
 DEFAULT_NUM_URLS = defaultdict(lambda: 3)
 DEFAULT_NUM_URLS["prediction-online-summarized-info"] = 7
@@ -436,7 +436,9 @@ def run(**kwargs) -> Tuple[str, Optional[str], Optional[Dict[str, Any]], Any]:
         if tool not in ALLOWED_TOOLS:
             raise ValueError(f"Tool {tool} is not supported.")
 
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
+
         if tool.startswith("prediction-online"):
             additional_information, counter_callback = fetch_additional_information(
                 prompt,

packages/valory/customs/prediction_request_embedding/prediction_sentence_embedding.py

@@ -80,8 +80,8 @@ def count_tokens(text: str, model: str) -> int:
     "prediction-sentence-embedding-bold",
 ]
 TOOL_TO_ENGINE = {
-    "prediction-sentence-embedding-conservative": "gpt-3.5-turbo",
-    "prediction-sentence-embedding-bold": "gpt-4",
+    "prediction-sentence-embedding-conservative": "gpt-3.5-turbo-0125",
+    "prediction-sentence-embedding-bold": "gpt-4-0125-preview",
 }
 
 
@@ -1040,7 +1040,7 @@ def fetch_additional_information(
     google_api_key: str,
     google_engine: str,
     nlp,
-    engine: str = "gpt-3.5-turbo",
+    engine: str = "gpt-4-0125-preview",
     temperature: float = 0.5,
     max_compl_tokens: int = 500,
 ) -> str:
@@ -1053,7 +1053,7 @@ def fetch_additional_information(
         google_api_key (str): The API key for the Google service.
         google_engine (str): The Google engine to be used.
         temperature (float): The temperature parameter for the engine.
-        engine (str): The openai engine. Defaults to "gpt-3.5-turbo".
+        engine (str): The openai engine. Defaults to "gpt-4-0125-preview".
         temperature (float): The temperature parameter for the engine. Defaults to 1.0.
         max_compl_tokens (int): The maximum number of tokens for the engine's response.
 
@@ -1142,7 +1142,8 @@ def run(**kwargs) -> Tuple[Optional[str], Any, Optional[Dict[str, Any]], Any]:
         nlp = spacy.load("en_core_web_md")
 
         # Get the LLM engine to be used
-        engine = TOOL_TO_ENGINE[tool]
+        engine = kwargs.get("model", TOOL_TO_ENGINE[tool])
+        print(f"ENGINE: {engine}")
 
         # Extract the event question from the prompt
         event_question = re.search(r"\"(.+?)\"", prompt).group(1)
@@ -1163,7 +1164,7 @@ def run(**kwargs) -> Tuple[Optional[str], Any, Optional[Dict[str, Any]], Any]:
         # Fetch additional information
         additional_information = fetch_additional_information(
             event_question=event_question,
-            engine="gpt-3.5-turbo",
+            engine="gpt-4-0125-preview",
             temperature=0.5,
             max_compl_tokens=max_compl_tokens,
             nlp=nlp,