letta icon indicating copy to clipboard operation
letta copied to clipboard

MemGPT endpoints / inference.memgpt.ai: Bad output: Broken json -- bad formed responses

Open quantumalchemy opened this issue 9 months ago โ€ข 0 comments

Describe the bug ยป Use the free MemGPT endpoints inference.memgpt.ai getting broken json -- bad formed responses: in CLI / and api model = memgpt-openai model_endpoint = https://inference.memgpt.ai model_endpoint_type = openai context_window = 16384

OUTPUT:

Creating new agent... Couldn't find human {human} in database, please run memgpt add human Couldn't find persona {persona} in database, please run memgpt add persona -> ๐Ÿค– Using persona profile: 'sam_pov' -> ๐Ÿง‘ Using human profile: 'basic' ๐ŸŽ‰ Created new agent 'CheerfulWombat' (id=3ff423f2-7556-487c-95e0-357cddec62e0)

Hit enter to begin (will request first MemGPT message) response = <Response [200]>, response.text = {"id":"chatcmpl-9KsAUJwZJzXRFyx2Rt9PmzjedYQxy","choices":[{"finish_reason":"tool_calls","index":0,"message":{"content":null,"tool_calls":[{"id":"call_zOQk8RFziX5v u96A3kAmdIYc","type":"function","function":{"arguments":"{"message":"Hello, Chad! Welcome. I'm Sam. It's great to meet you. What brings you here today?"}","name":"send_message"}}],"role":"assistant","function_call":null},"logprobs":null}],"created":"2024-05-03T18:43:30Z","model":"gpt-4-turbo-2024-04-09"," system_fingerprint":"fp_ea6eb70039","object":"chat.completion","usage":{"completion_tokens":36,"prompt_tokens":2302,"total_tokens":2338}} ๐Ÿ’ญ None ๐Ÿค– Hello, Chad! Welcome. I'm Sam. It's great to meet you. What brings you here today?

Enter your message: hello

response = <Response [200]>, response.text = {"id":"chatcmpl-9KsAfLhomNaEugXEJPP5yTYURHpFX","choices":[{"finish_reason":"tool_calls","index":0,"message":{"content":null,"tool_calls":[{"id":"call_OXtdy7amyo0F 0wxzYN02kiCK","type":"function","function":{"arguments":"{"message":"Hi! How's your day going?"}","name":"send_message"}}],"role":"assistant","function_call":null},"logprobs":null}],"created":"2024-05-03T18:43:41Z","model":"gpt-4-turbo-2024-04-09"," system_fingerprint":"fp_ea6eb70039","object":"chat.completion","usage":{"completion_tokens":21,"prompt_tokens":2419,"total_tokens":2440}} ๐Ÿ’ญ None

Please describe your setup

  • [ ] How did you install memgpt?
    • pip install pymemgpt? pip install pymemgpt-nightly? git clone?
  • [ ] Describe your setup
    • What's your OS (Windows/MacOS/Linux)?
    • How are you running memgpt? (cmd.exe/Powershell/Anaconda Shell/Terminal)

MemGPT Config [defaults] preset = memgpt_chat persona = sam_pov human = basic

[model] model = memgpt-openai model_endpoint = https://inference.memgpt.ai model_endpoint_type = openai context_window = 16384

[embedding] embedding_endpoint_type = hugging-face embedding_endpoint = https://embeddings.memgpt.ai embedding_model = BAAI/bge-large-en-v1.5 embedding_dim = 1024 embedding_chunk_size = 300

[archival_storage] type = chroma path = /root/.memgpt/chroma

[recall_storage] type = sqlite path = /root/.memgpt

[metadata_storage] type = sqlite path = /root/.memgpt

[version] memgpt_version = 0.3.13

[client] anon_clientid = 00000000000000000000867a4627fd44


https://inference.memgpt.ai

quantumalchemy avatar May 03 '24 18:05 quantumalchemy