You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If you're not using OpenAI, please provide additional information on your local LLM setup:
Local LLM details
If you are trying to run MemGPT with local LLMs, please provide the following information:
with tools : read_from_text_file, message_chatgpt, http_request, append_to_text_file
the result is
NFO: 127.0.0.1:33496 - "GET /api/agents/13190d52-f321-427f-a43c-f166225ea419/memory HTTP/1.1" 200 OK
INFO: 127.0.0.1:60924 - "POST /api/agents/13190d52-f321-427f-a43c-f166225ea419/messages HTTP/1.1" 200 OK
{"type": "user_message", "message": "can you give me the content of this page : https://spoggy-test2.solidcommunity.net/public/brains/Chateau_des_Robots/", "time": "2024-06-13 11:08:56 PM CEST+0200"}
{'id': UUID('c2460d8c-ec11-486d-add3-89a0809f49a8'), 'user_id': UUID('00000000-0000-0000-0000-000000000000'), 'agent_id': UUID('13190d52-f321-427f-a43c-f166225ea419'), 'text': '{"type": "user_message", "message": "can you give me the content of this page : https://spoggy-test2.solidcommunity.net/public/brains/Chateau_des_Robots/", "time": "2024-06-13 11:08:56 PM CEST+0200"}', 'model': 'llama3-8b-8192', 'created_at': datetime.datetime(2024, 6, 13, 21, 8, 56, 557883, tzinfo=datetime.timezone.utc), 'role': 'user', 'name': None, 'embedding': None, 'embedding_dim': None, 'embedding_model': None, 'tool_calls': None, 'tool_call_id': None}
2024-06-13T21:08:56.557883+00:00
handling asyncio exception {'message': 'Future exception was never retrieved', 'exception': LocalLLMError('Failed to parse JSON from local LLM response - error: Failed to decode JSON from LLM output:\n\n{\n "function":I\'m not programmed to provide external content. - error\nFailed to decode valid MemGPT JSON from LLM output:\n=====\n\n{\n "function":I\'m not programmed to provide external content.\n====='), 'future': <Future finished exception=LocalLLMError('Failed to parse JSON from local LLM response - error: Failed to decode JSON from LLM output:\n\n{\n "function":I\'m not programmed to provide external content. - error\nFailed to decode valid MemGPT JSON from LLM output:\n=====\n\n{\n "function":I\'m not programmed to provide external content.\n=====')>}
then it does not use the function http_request whrereas it is configured
With git clone and using the CLI
💭 Ah, bonjour! Bienvenue dans notre conversation. Je m'appelle Sam, et je suis ravi de vous rencontrer.
🤖 Ah, bonjour! Je suis Sam, et je suis ravi de vous rencontrer. Je vous répondrai en français si ça vous convient.
> Enter your message: peux-tu me donner le contenu de cette page https://spoggy-test2.solidcommunity.net/public/brains/Chateau_des_Robots/ ?
An exception occurred when running agent.step():
Traceback (most recent call last):
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/llm_chat_completion_wrappers/llama3.py", line 319, in output_to_chat_completion_response
function_json_output = clean_json(raw_llm_output)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/json_parser.py", line 202, in clean_json
raise LLMJSONParsingError(f"Failed to decode valid MemGPT JSON from LLM output:\n=====\n{raw_llm_output}\n=====")
memgpt.errors.LLMJSONParsingError: Failed to decode valid MemGPT JSON from LLM output:
=====
{
"function":I apologize, it seems like there was a premature termination of the program. As Sam, I would like to clarify that I'm not programmed to access external websites or provide the content you're requesting. My capabilities are limited to conversing with you as a person. However, I can certainly entertain you with a conversation!
=====
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 189, in get_chat_completion
chat_completion_result = llm_wrapper.output_to_chat_completion_response(result, first_message=first_message)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/llm_chat_completion_wrappers/llama3.py", line 321, in output_to_chat_completion_response
raise Exception(f"Failed to decode JSON from LLM output:\n{raw_llm_output} - error\n{str(e)}")
Exception: Failed to decode JSON from LLM output:
{
"function":I apologize, it seems like there was a premature termination of the program. As Sam, I would like to clarify that I'm not programmed to access external websites or provide the content you're requesting. My capabilities are limited to conversing with you as a person. However, I can certainly entertain you with a conversation! - error
Failed to decode valid MemGPT JSON from LLM output:
=====
{
"function":I apologize, it seems like there was a premature termination of the program. As Sam, I would like to clarify that I'm not programmed to access external websites or provide the content you're requesting. My capabilities are limited to conversing with you as a person. However, I can certainly entertain you with a conversation!
=====
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/main.py", line 408, in run_agent_loop
new_messages, user_message, skip_next_user_input = process_agent_step(user_message, no_verify)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/main.py", line 377, in process_agent_step
new_messages, heartbeat_request, function_failed, token_warning, tokens_accumulated = memgpt_agent.step(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 823, in step
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 751, in step
response = self._get_ai_reply(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 456, in _get_ai_reply
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 431, in _get_ai_reply
response = create(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 133, in wrapper
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 106, in wrapper
return func(*args, **kwargs)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 324, in create
return get_chat_completion(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 194, in get_chat_completion
raise LocalLLMError(f"Failed to parse JSON from local LLM response - error: {str(e)}")
memgpt.errors.LocalLLMError: Failed to parse JSON from local LLM response - error: Failed to decode JSON from LLM output:
{
"function":I apologize, it seems like there was a premature termination of the program. As Sam, I would like to clarify that I'm not programmed to access external websites or provide the content you're requesting. My capabilities are limited to conversing with you as a person. However, I can certainly entertain you with a conversation! - error
Failed to decode valid MemGPT JSON from LLM output:
=====
{
"function":I apologize, it seems like there was a premature termination of the program. As Sam, I would like to clarify that I'm not programmed to access external websites or provide the content you're requesting. My capabilities are limited to conversing with you as a person. However, I can certainly entertain you with a conversation!
=====
? Retry agent.step()? (Y/n)
removing error handling in clean_json
Traceback (most recent call last):
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 189, in get_chat_completion
chat_completion_result = llm_wrapper.output_to_chat_completion_response(result, first_message=first_message)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/llm_chat_completion_wrappers/llama3.py", line 321, in output_to_chat_completion_response
raise Exception(f"Failed to decode JSON from LLM output:\n{raw_llm_output} - error\n{str(e)}")
Exception: Failed to decode JSON from LLM output:
{
"function": "send_message",{
"function": "send_message",
"params": {
"inner_thoughts": "Initialization complete. Let's begin our conversation.",
"message": "Nice to meet you, Chad!"
}
} - error
Expecting property name enclosed in double quotes: line 3 column 30 (char 32)
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/main.py", line 408, in run_agent_loop
new_messages, user_message, skip_next_user_input = process_agent_step(user_message, no_verify)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/main.py", line 377, in process_agent_step
new_messages, heartbeat_request, function_failed, token_warning, tokens_accumulated = memgpt_agent.step(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 823, in step
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 738, in step
response = self._get_ai_reply(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 456, in _get_ai_reply
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/agent.py", line 431, in _get_ai_reply
response = create(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 133, in wrapper
raise e
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 106, in wrapper
return func(*args, **kwargs)
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/llm_api/llm_api_tools.py", line 324, in create
return get_chat_completion(
File "/home/smag/dev/quest/bot/.venv/lib/python3.10/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 194, in get_chat_completion
raise LocalLLMError(f"Failed to parse JSON from local LLM response - error: {str(e)}")
memgpt.errors.LocalLLMError: Failed to parse JSON from local LLM response - error: Failed to decode JSON from LLM output:
{
"function": "send_message",{
"function": "send_message",
"params": {
"inner_thoughts": "Initialization complete. Let's begin our conversation.",
"message": "Nice to meet you, Chad!"
}
} - error
Expecting property name enclosed in double quotes: line 3 column 30 (char 32)
? Retry agent.step()? (Y/n)
The text was updated successfully, but these errors were encountered:
Describe the bug
Groq does not work with function, the clean_json does not extract the functions
might be related to #1309 and #1257
Please describe your setup
pip install pymemgpt
and/orpip install pymemgpt-nightly
Screenshots
If applicable, add screenshots to help explain your problem.
Additional context
persona : sam_pov
user: basic Chad
MemGPT Config
``
[defaults]
preset = memgpt_chat
persona = sam_pov
human = basic
[model]
model = llama3-8b-8192
model_endpoint = https://api.groq.com/openai
model_endpoint_type = groq
model_wrapper = llama3
context_window = 8192
[embedding]
embedding_endpoint_type = local
embedding_model = BAAI/bge-small-en-v1.5
embedding_dim = 384
embedding_chunk_size = 300
[archival_storage]
type = chroma
path = /home/smag/.memgpt/chroma
[recall_storage]
type = sqlite
path = /home/smag/.memgpt
[metadata_storage]
type = sqlite
path = /home/smag/.memgpt
[version]
memgpt_version = 0.3.17
[client]
anon_clientid = 00000000-0000-0000-0000-000000000000
``
If you're not using OpenAI, please provide additional information on your local LLM setup:
Local LLM details
If you are trying to run MemGPT with local LLMs, please provide the following information:
with tools : read_from_text_file, message_chatgpt, http_request, append_to_text_file
the result is
then it does not use the function http_request whrereas it is configured
With git clone and using the CLI
removing error handling in clean_json
The text was updated successfully, but these errors were encountered: