cpacker
commented
5 months ago Hey @cleesmith thanks for opening the issue - our free endpoint was down earlier today but is back online (https://status.memgpt.ai/status/all). Can you confirm if it's working now?
In the future we can try to revise the error messaging to be a little more communicative - eg in this case, it might have helped if we returned / printed an error that looked like:
Error: can't reach the MemGPT hosted endpoint. The MemGPT hosted endpoint may be offline - you can check its current status at https://status.memgpt.ai
cleesmith
sarahwooders
Describe the bug It does not work
Please describe your setup
Screenshots ... in the Terminal: memgpt run
? Would you like to select an existing agent? Yes ? Select agent: UnderstandingLighthouse
🔁 Using existing agent UnderstandingLighthouse
Hit enter to begin (will request first MemGPT message)
An exception occurred when running agent.step(): Traceback (most recent call last): File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/main.py", line 356, in run_agent_loop new_messages, user_message, skip_next_user_input = process_agent_step(user_message, no_verify) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/main.py", line 332, in process_agent_step new_messages, heartbeat_request, function_failed, token_warning, tokens_accumulated = memgpt_agent.step( ^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/agent.py", line 676, in step raise e File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/agent.py", line 596, in step response = self._get_ai_reply( ^^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/agent.py", line 362, in _get_ai_reply raise e File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/agent.py", line 342, in _get_ai_reply response = create( ^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 373, in wrapper raise e File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 346, in wrapper return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 465, in create return get_chat_completion( ^^^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 159, in get_chat_completion result, usage = get_vllm_completion(endpoint, auth_type, auth_key, model, prompt, context_window, user) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/cleesmith/anaconda3/envs/memgpt/lib/python3.11/site-packages/memgpt/local_llm/vllm/api.py", line 46, in get_vllm_completion raise Exception( Exception: API call got non-200 response code (code=521, msg=<!DOCTYPE html>
Web server is down Error code 521
Browser
WorkingCloudflare
WorkingHost
ErrorWhat happened?
The web server is not returning a connection. As a result, the web page is not displaying.
What can I do?
If you are a visitor of this website:
Please try again in a few minutes.
If you are the owner of this website:
Contact your hosting provider letting them know your web server is not responding. Additional troubleshooting information.
Cloudflare Ray ID: 85044a89daaf7d20 • Your IP: 2603:7080:5d4a:d500:f836:2a4b:1176:b94c • Performance & security by Cloudflare
) for address: https://api.memgpt.ai/v1/completions. Make sure that the vLLM server is running and reachable at https://api.memgpt.ai/v1/completions. ? Retry agent.step()? No
Cancelled by user
Finished.
...... It seems to always try to go to: https://api.memgpt.ai/v1/completions (mothership?) ... regardless of the "configure" being set to: openai or local. Why? It's never a good impresion when the simple quickstart does not work.