Show details for query tokens (#1392)

embedchain/llm/nvidia.py

@@ -1,6 +1,6 @@
 import os
 from collections.abc import Iterable
-from typing import Optional, Union
+from typing import Any, Optional, Union
 
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.stdout import StdOutCallbackHandler
@@ -25,8 +25,27 @@ class NvidiaLlm(BaseLlm):
         if not self.config.api_key and "NVIDIA_API_KEY" not in os.environ:
             raise ValueError("Please set the NVIDIA_API_KEY environment variable or pass it in the config.")
 
-    def get_llm_model_answer(self, prompt):
-        return self._get_answer(prompt=prompt, config=self.config)
+    def get_llm_model_answer(self, prompt) -> tuple[str, Optional[dict[str, Any]]]:
+        if self.config.token_usage:
+            response, token_info = self._get_answer(prompt, self.config)
+            model_name = "nvidia/" + self.config.model
+            if model_name not in self.config.model_pricing_map:
+                raise ValueError(
+                    f"Model {model_name} not found in `model_prices_and_context_window.json`. \
+                    You can disable token usage by setting `token_usage` to False."
+                )
+            total_cost = (
+                self.config.model_pricing_map[model_name]["input_cost_per_token"] * token_info["input_tokens"]
+            ) + self.config.model_pricing_map[model_name]["output_cost_per_token"] * token_info["output_tokens"]
+            response_token_info = {
+                "prompt_tokens": token_info["input_tokens"],
+                "completion_tokens": token_info["output_tokens"],
+                "total_tokens": token_info["input_tokens"] + token_info["output_tokens"],
+                "total_cost": round(total_cost, 10),
+                "cost_currency": "USD",
+            }
+            return response, response_token_info
+        return self._get_answer(prompt, self.config)
 
     @staticmethod
     def _get_answer(prompt: str, config: BaseLlmConfig) -> Union[str, Iterable]:
@@ -43,4 +62,7 @@ class NvidiaLlm(BaseLlm):
         if labels:
             params["labels"] = labels
         llm = ChatNVIDIA(**params, callback_manager=CallbackManager(callback_manager))
-        return llm.invoke(prompt).content if labels is None else llm.invoke(prompt, labels=labels).content
+        chat_response = llm.invoke(prompt) if labels is None else llm.invoke(prompt, labels=labels)
+        if config.token_usage:
+            return chat_response.content, chat_response.response_metadata["token_usage"]
+        return chat_response.content