mem0-mcp-selfhosted
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MEM0_HOST | No | Host for SSE/HTTP transports. | 0.0.0.0 |
| MEM0_PORT | No | Port for SSE/HTTP transports. | 8081 |
| MEM0_LLM_URL | No | Ollama base URL for the main LLM. Cascades: MEM0_LLM_URL -> MEM0_OLLAMA_URL -> http://localhost:11434. Only used when MEM0_LLM_PROVIDER=ollama. | http://localhost:11434 |
| MEM0_USER_ID | No | Default user ID for memory scoping. Used when a tool's user_id parameter is not provided. | user |
| MEM0_PROVIDER | No | Top-level LLM provider ('anthropic' or 'ollama'). Cascades to MEM0_LLM_PROVIDER and MEM0_GRAPH_LLM_PROVIDER when those are not set. Does not affect MEM0_EMBED_PROVIDER. | anthropic |
| GOOGLE_API_KEY | No | Google API key (required for the gemini/gemini_split graph providers). | |
| MEM0_EMBED_URL | No | Ollama URL for embeddings. Cascades: MEM0_EMBED_URL -> MEM0_OLLAMA_URL -> http://localhost:11434. | http://localhost:11434 |
| MEM0_LLM_MODEL | No | Model for the selected LLM provider. Defaults to claude-opus-4-6 for Anthropic and qwen3:14b for Ollama. | claude-opus-4-6 |
| MEM0_LOG_LEVEL | No | Logging level ('DEBUG', 'INFO', 'WARNING', 'ERROR'). | INFO |
| MEM0_NEO4J_URL | No | Neo4j Bolt endpoint. | bolt://127.0.0.1:7687 |
| MEM0_TRANSPORT | No | Transport: 'stdio', 'sse', or 'streamable-http'. | stdio |
| MEM0_COLLECTION | No | Qdrant collection name. | mem0_mcp_selfhosted |
| MEM0_EMBED_DIMS | No | Embedding vector dimensions. | 1024 |
| MEM0_NEO4J_USER | No | Neo4j username. | neo4j |
| MEM0_OLLAMA_URL | No | Shared Ollama base URL. Cascades to MEM0_LLM_URL, MEM0_EMBED_URL, and MEM0_GRAPH_LLM_URL when those are not set. | http://localhost:11434 |
| MEM0_QDRANT_URL | No | Qdrant REST API URL. | http://localhost:6333 |
| MEM0_EMBED_MODEL | No | Embedding model name. | bge-m3 |
| MEM0_OAT_HEADERS | No | OAT identity headers: 'auto' or 'none'. | auto |
| ANTHROPIC_API_KEY | No | Standard Anthropic API key (priority 3 in the token fallback chain). Only useful for non-Claude-Code deployments (Docker, CI, standalone). | |
| MEM0_ENABLE_GRAPH | No | Enable graph memory (entity extraction to Neo4j). Disabled by default to protect your Claude quota. | false |
| MEM0_LLM_PROVIDER | No | Main LLM provider: 'anthropic' or 'ollama'. Inherits from MEM0_PROVIDER when not set. | anthropic |
| MEM0_OLLAMA_THINK | No | Set to true to re-enable qwen3 thinking mode (disabled by default to prevent thinking + format:"json" collision). | false |
| MEM0_GRAPH_LLM_URL | No | Ollama base URL for graph LLM. Cascades: MEM0_GRAPH_LLM_URL -> MEM0_LLM_URL -> MEM0_OLLAMA_URL -> http://localhost:11434. | http://localhost:11434 |
| MEM0_EMBED_PROVIDER | No | Embedding provider ('ollama' or 'openai'). | ollama |
| MEM0_LLM_MAX_TOKENS | No | Max tokens for LLM responses (Anthropic only). | 16384 |
| MEM0_NEO4J_DATABASE | No | Neo4j database name (multi-database setups). | |
| MEM0_NEO4J_PASSWORD | No | Neo4j password. | mem0graph |
| MEM0_QDRANT_API_KEY | No | Qdrant API key (for Qdrant Cloud). | |
| MEM0_QDRANT_ON_DISK | No | Store vectors on disk (reduces RAM, slower search). | false |
| MEM0_QDRANT_TIMEOUT | No | Qdrant REST API timeout in seconds (e.g., 30). Only set if you hit ReadTimeout during collection operations. | |
| MEM0_ANTHROPIC_TOKEN | No | Anthropic OAT or API token (highest priority in the token fallback chain). | |
| MEM0_GRAPH_LLM_MODEL | No | Graph model. Inherits MEM0_LLM_MODEL for anthropic/ollama; defaults to gemini-2.5-flash-lite for gemini/gemini_split. | |
| MEM0_GRAPH_THRESHOLD | No | Embedding similarity threshold for node matching. | 0.7 |
| MEM0_HISTORY_DB_PATH | No | SQLite path for memory change history. | |
| MEM0_NEO4J_BASE_LABEL | No | Custom Neo4j base label for node type grouping. | |
| MEM0_OLLAMA_KEEP_ALIVE | No | How long Ollama keeps the model in VRAM between calls (e.g., 1h, 5m). Prevents model unload during multi-call graph pipelines. | 30m |
| MEM0_GRAPH_LLM_PROVIDER | No | Graph LLM provider ('anthropic', 'anthropic_oat', 'ollama', 'gemini', 'gemini_split'). Inherits from MEM0_PROVIDER when not set. | anthropic |
| MEM0_GRAPH_CONTRADICTION_LLM_MODEL | No | Contradiction model in gemini_split mode. Defaults to claude-opus-4-6 for anthropic/anthropic_oat providers; inherits MEM0_LLM_MODEL for others. | |
| MEM0_OAT_REFRESH_THRESHOLD_SECONDS | No | Seconds before expiry to trigger proactive OAT token refresh. | 1800 |
| MEM0_GRAPH_CONTRADICTION_LLM_PROVIDER | No | Contradiction LLM provider in gemini_split mode ('anthropic', 'anthropic_oat', 'ollama'). | anthropic |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
This server cannot be deployed
Maintenance
ActivityActive
ResponsivenessNo issues