- _get_all_from_vector_store: top_k -> limit (2 missed occurrences)
- test_empty_llm_response_fact_extraction (sync+async): add
db.get_last_messages mock returning [], set custom_instructions=None,
fix log assertion to use record.message not record.msg
- test_thinking_tags (sync+async): add embed_batch mock, db mock
with get_last_messages=[], set custom_instructions=None
Root cause: v3 pipeline calls self.db.get_last_messages() and
self.embedding_model.embed_batch() which weren't mocked, causing
the pipeline to silently fail before reaching the LLM call.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The async context manager tests that used asyncio were deleted in the
previous commit. The import is no longer needed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tests were written for the old two-pass extraction pipeline (extract
facts + merge with ADD/UPDATE/DELETE). V3 uses single-pass ADD-only
extraction with one LLM call returning {"memory": [...]}.
Changes by category:
Old pipeline tests (13 failures):
- Delete TestHallucinatedIdGuard entirely (tests UPDATE/DELETE ID
resolution that no longer exists)
- Update error log assertions to match v3 log messages
- Change LLM mock from 2-call side_effect to 1-call return_value
- Update vllm thinking tag tests for single-call format
ensure_json_instruction removed (3):
- Delete source-inspection tests (function still exists in utils
but is no longer called from _add_to_vector_store)
UTC timestamp normalization removed (2):
- Update expected timestamps to stored-as-is (no UTC conversion)
- Remove _normalize_iso_timestamp_to_utc import
_search_vector_store signature: top_k -> limit (2):
- Update test calls to use limit= parameter
Context manager + close/db removed (6):
- Delete tests for __enter__/__exit__/close/db that were removed
Vector store BM25 config (3):
- Qdrant: add sparse_vectors_config to create_col assertion,
named vector format {"": vector} in update assertion
- MongoDB: assert_any_call for both vector + text search indexes
Misc (6):
- test_search_handles_incomplete_payloads: expect 1 result (v3
filters entries without 'data' key)
- Embedding cache tests: update for embed_batch (1 call) vs
individual embed (was 2 calls)
- Graph reset tests: delete (graph support removed)
- use_azure_credential: update expected sensitivity to True
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Remove test_neo4j_cypher_syntax.py and test_graph_delete_docker.py
which reference deleted graph_memory module.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Strip graph_memory patches, GraphStoreFactory mocks, graph_store config,
enable_graph parametrize, _add_to_graph assertions, and "relations"
assertions from all test fixtures and test functions. Graph store has
been removed from the codebase.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Restore LMStudioEmbedder, LMStudioLLM, DeepSeekLLM in TS factory.ts
(accidentally removed during graph store cleanup — unrelated to graphs)
- Add langchain>=0.1.0 to Python extras dep group (was transitively
provided by langchain-neo4j in the removed graph group, still needed
by mem0/llms/langchain.py and mem0/embeddings/langchain.py)
- Add pytest.importorskip("langchain") to test_langchain.py for graceful
skip when langchain isn't installed
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
These test files import mem0.memory.graph_memory, kuzu_memory,
memgraph_memory, apache_age_memory, and mem0.graphs.neptune which
were deleted in the graph store removal commit.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove deprecated custom_update_memory_prompt from MemoryConfig entirely
- Add MAX_BATCH=100 chunking guard to embed_batch in openai.py and azure_openai.py
- Move all inline imports (mem0.utils.*) to top-level in main.py
- Remove custom_update_memory_prompt from test fixture
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace the 2-LLM-call add pipeline with a single-pass additive extraction
using ADDITIVE_EXTRACTION_PROMPT. Memories now accumulate (ADD-only) instead
of being updated/deleted during extraction.
Search pipeline upgraded to hybrid scoring combining three signals:
- Semantic similarity (vector search)
- BM25 keyword matching (native per vector store, 15 stores supported)
- Entity boost (spaCy NER with entity collection linking)
Combined score = (semantic + bm25 + entity_boost) / max_possible, where
max_possible adapts based on which signals are active.
Key changes:
- Add spaCy-based lemmatization for BM25 keyword search
- Add spaCy-based entity extraction (PROPER, QUOTED, COMPOUND, NOUN types)
- Add entity store as second vector collection ({collection}_entities)
- Add native keyword_search() to 15 vector store adapters
- Add batch embedding support (embed_batch) for OpenAI and Azure OpenAI
- Add message persistence in SQLite (rolling window of 10 per scope)
- Add additive scoring with adaptive normalization
- Add observation_date parameter to add()
- Add custom_instructions config field
- Default search threshold changed to 0.1
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
kartik-mem0