llm_server/llm/vllm/tokenize.py

@@ -29,6 +29,7 @@ def tokenize(prompt: str, backend_url: str) -> int:
     # Define a function to send a chunk to the server
     def send_chunk(chunk):
         try:
+            XXX
             r = requests.post(f'{backend_url}/tokenize', json={'input': chunk}, verify=opts.verify_ssl, timeout=opts.backend_generate_request_timeout)
             j = r.json()
             return j['length']