Describe the bug
run amemgym benchmark using memmachine semantic memory.
after running benchmark, look at docker logs for memmachine.
although no issues observed in restapiv2 responses, found exception in docker logs.
2026-02-26 22:55:20,531 [ERROR] memmachine.common.language_model.openai_chat_completions_language_model - [call uuid: 7e4c4c3c-27f9-4710-b107-301cdce72540] Giving up generating response due to non-retryable LengthFinishReasonError
...
File "/app/.venv/lib/python3.12/site-packages/openai/lib/_parsing/_completions.py", line 100, in parse_chat_completion
raise LengthFinishReasonError(completion=chat_completion)
openai.LengthFinishReasonError: Could not parse response content as the length limit was reached - CompletionUsage(completion_tokens=16384, prompt_tokens=8034, total_tokens=24418, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=2944))
2026-02-26 22:55:20,546 [ERROR] memmachine.semantic_memory.semantic_ingestion - Failed to process message 168 for semantic type profile
Traceback (most recent call last):
File "/app/.venv/lib/python3.12/site-packages/memmachine/common/language_model/openai_chat_completions_language_model.py", line 146, in generate_parsed_response
response = await self._client.with_options(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...
Steps to reproduce
how to run amemgym benchmark is in this message:
https://memverge.slack.com/archives/C0A0Z69UC8L/p1771976896651179
Expected behavior
no errors
Environment
build is 2/25 main 46d7a05
Additional context
01-logs.txt
Describe the bug
run amemgym benchmark using memmachine semantic memory.
after running benchmark, look at docker logs for memmachine.
although no issues observed in restapiv2 responses, found exception in docker logs.
2026-02-26 22:55:20,531 [ERROR] memmachine.common.language_model.openai_chat_completions_language_model - [call uuid: 7e4c4c3c-27f9-4710-b107-301cdce72540] Giving up generating response due to non-retryable LengthFinishReasonError
...
File "/app/.venv/lib/python3.12/site-packages/openai/lib/_parsing/_completions.py", line 100, in parse_chat_completion
raise LengthFinishReasonError(completion=chat_completion)
openai.LengthFinishReasonError: Could not parse response content as the length limit was reached - CompletionUsage(completion_tokens=16384, prompt_tokens=8034, total_tokens=24418, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=2944))
2026-02-26 22:55:20,546 [ERROR] memmachine.semantic_memory.semantic_ingestion - Failed to process message 168 for semantic type profile
Traceback (most recent call last):
File "/app/.venv/lib/python3.12/site-packages/memmachine/common/language_model/openai_chat_completions_language_model.py", line 146, in generate_parsed_response
response = await self._client.with_options(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...
Steps to reproduce
how to run amemgym benchmark is in this message:
https://memverge.slack.com/archives/C0A0Z69UC8L/p1771976896651179
Expected behavior
no errors
Environment
build is 2/25 main 46d7a05
Additional context
01-logs.txt