feat: allow boto3 to use its native credential finding functionality (#1536)

embedchain/docs/components/llms.mdx

@@ -708,8 +708,6 @@ embedder:
 import os
 from embedchain import App
 
-os.environ["AWS_ACCESS_KEY_ID"] = "xxx"
-os.environ["AWS_SECRET_ACCESS_KEY"] = "xxx"
 os.environ["AWS_REGION"] = "us-west-2"
 
 app = App.from_config(config_path="config.yaml")

mem0/llms/aws_bedrock.py

@@ -1,5 +1,5 @@
+
 import json
-import os
 from typing import Any, Dict, List, Optional
 
 try:
@@ -17,12 +17,7 @@ class AWSBedrockLLM(LLMBase):
 
         if not self.config.model:
             self.config.model="anthropic.claude-3-5-sonnet-20240620-v1:0"
-        self.client = boto3.client(
-            "bedrock-runtime",
-            region_name=os.environ.get("AWS_REGION"),
-            aws_access_key_id=os.environ.get("AWS_ACCESS_KEY"),
-            aws_secret_access_key=os.environ.get("AWS_SECRET_ACCESS_KEY"),
-        )
+        self.client = boto3.client("bedrock-runtime")
         self.model_kwargs = {
             "temperature": self.config.temperature,
             "max_tokens_to_sample": self.config.max_tokens,