mem0 · Issues· 759 open

AWS Bedrock: non-tool Amazon Nova path sends Converse but parses with the invoke_model parser, so generate_response returns "Error parsing response"

Upgrade mem0-ts to OpenAI SDK v5

bug(server): no rate limiting on LLM-calling endpoints (/memories, /search, /generate-instructions)

mem0 TypeScript ignores env-specified proxy settings (http_proxy, HTTPS_PROXY) when talking to model providers

Truncated extraction output (max_tokens hit mid-JSON) drops all facts, including the ones already written

docs: Python multimodal examples recreate clients with vision disabled

ADD-only extraction in v3 may surface stale/contradictory facts for time-sensitive attributes

BM25 keyword search and entity extraction are hardcoded to English

bug(oss): ConfigManager injects OpenAI's baseURL and model into every other LLM provider, so documented non-OpenAI configs are sent to api.openai.com

bug(vector_stores/azure-ai-search): telemetry helpers operate on the memory index — a user's memory is silently re-scoped to the telemetry id

Claude Code plugin reports "API key missing" after a successful `mem0 login` — `plugin_sync` only updates pre-existing entries

Windows: Cursor plugin 0.2.13 spawns hung Git Bash (mintty) windows on Read hooks

Deepseek Harness: canceled recall marks undelivered memories as seen

[Bug] AWS Bedrock: legacy amazon (Titan) responses always parse to '' — _parse_response reads a 'completion' field Titan never returns

PGVector destructor closes externally supplied connection_pool