use archfc hosted on aws

model_server/src/cli.py

@@ -129,18 +129,18 @@ def stop_server():
 
     pid_file = get_pid_file()
     if os.path.exists(pid_file):
-        logger.info("PID file found, shutting down the server.")
+        logger.info("pid file found, shutting down the server.")
         # read pid from file
         with open(pid_file, "r") as f:
             pid = int(f.read())
-            logger.info(f"Killing model server {pid}")
+            logger.info(f"killing model server {pid}")
             try:
                 os.kill(pid, signal.SIGKILL)
             except ProcessLookupError:
-                logger.info(f"Process {pid} not found")
+                logger.info(f"process {pid} not found")
         os.remove(pid_file)
     else:
-        logger.info("No PID file found, server is not running.")
+        logger.info("no pid file found, server is not running.")
 
 
 def restart_server(port=51000, foreground=False):

model_server/src/commons/globals.py

@@ -15,7 +15,7 @@ logger = get_model_server_logger()
 
 
 # Define the client
-ARCH_ENDPOINT = os.getenv("ARCH_ENDPOINT", "https://api.fc.archgw.com/v1")
+ARCH_ENDPOINT = os.getenv("ARCH_ENDPOINT", "https://archfc.katanemo.dev/v1")
 ARCH_API_KEY = "EMPTY"
 ARCH_CLIENT = OpenAI(base_url=ARCH_ENDPOINT, api_key=ARCH_API_KEY)
 

model_server/src/main.py

@@ -4,7 +4,7 @@ import time
 import logging
 import src.commons.utils as utils
 
-from src.commons.globals import handler_map
+from src.commons.globals import ARCH_ENDPOINT, handler_map
 from src.core.utils.model_utils import (
     ChatMessage,
     ChatCompletionResponse,
@@ -51,6 +51,8 @@ logging.getLogger("opentelemetry.exporter.otlp.proto.grpc.exporter").setLevel(
 app = FastAPI()
 FastAPIInstrumentor().instrument_app(app)
 
+logger.info(f"using archfc endpoint: {ARCH_ENDPOINT}")
+
 
 @app.get("/healthz")
 async def healthz():

tests/rest/api_model_server.rest

@@ -1,6 +1,5 @@
 @model_server_endpoint = http://localhost:51000
-@archfc_endpoint = https://api.fc.archgw.com
-
+@archfc_endpoint = https://archfc.katanemo.dev
 
 ### talk to function calling endpoint
 POST {{model_server_endpoint}}/function_calling HTTP/1.1
@@ -119,7 +118,7 @@ Content-Type: application/json
 }
 
 ### talk to Arch-Intent directly for completion
-POST {{archfc_endpoint}}/v1/chat/completions HTTP/1.1
+POST {{{{archfc_endpoint}}}}/v1/chat/completions HTTP/1.1
 Content-Type: application/json
 
 {

tests/rest/insurance_agent.rest

@@ -1,5 +1,5 @@
 @model_server_endpoint = http://localhost:51000
-@archfc_endpoint = https://api.fc.archgw.com
+@archfc_endpoint = https://archfc.katanemo.dev
 
 ### multi turn conversation with intent, except parameter gathering
 
@@ -55,7 +55,7 @@ Content-Type: application/json
   ]
 }
 ### talk to Arch-Intent directly for completion
-POST https://api.fc.archgw.com/v1/chat/completions HTTP/1.1
+POST https://archfc.katanemo.dev/v1/chat/completions HTTP/1.1
 Content-Type: application/json
 
 {
@@ -126,7 +126,7 @@ Content-Type: application/json
   ]
 }
 ### talk to Arch-Intent directly for completion, expect No
-POST https://api.fc.archgw.com/v1/chat/completions HTTP/1.1
+POST https://archfc.katanemo.dev/v1/chat/completions HTTP/1.1
 Content-Type: application/json
 
 {

tests/rest/network_agent.rest

@@ -1,5 +1,5 @@
 @model_server_endpoint = http://localhost:51000
-@archfc_endpoint = https://api.fc.archgw.com
+@archfc_endpoint = https://archfc.katanemo.dev
 
 ### single turn function calling all parameters insurance agent summary