v2.0.15
Mem0 Python SDK (v2.0.15)
Changed 1
- LLMReranker's default model is now gpt-5-mini (was gpt-4o-mini)
Fixed 4
- delete_all() now paginates through the vector store in batches of 1000 instead of listing once, preventing memories from being silently left behind in accounts with more memories than a single page
- Cap Supabase search()/list() top_k at the vecs query limit of 1000 instead of erroring
- Fix Supabase col_info() crash by reading collection attributes directly instead of calling the removed describe() method
- Set size on Elasticsearch KNN search queries so results respect top_k instead of being capped at Elasticsearch's default of 10 hits
Bug Fixes:
- Core:
delete_all()now paginates through the vector store in batches of 1000 instead of listing once, so accounts with more memories than a single page (most vector stores default to ~100) had the remainder silently left behind (#6636) - Vector Stores: Cap Supabase
search()/list()top_kat thevecsquery limit of 1000 instead of erroring, and fix acol_info()crash by reading collection attributes directly instead of calling the removeddescribe()method (#6695) - Vector Stores: Set
sizeon Elasticsearch KNN search queries, so results respecttop_kinstead of being capped at Elasticsearch's default of 10 hits (#5910)
Changes:
- Rerankers:
LLMReranker's default model is nowgpt-5-mini(wasgpt-4o-mini) (#6703)