NewCost control: a token allowance a month for each model choice — when one runs out, the cheaper one answers.See how it works
memtro
Start for free
Client

Memtro for LibreChat

Add Memtro to librechat.yaml as a custom endpoint and as an MCP server, for the whole team or per person.

LibreChat takes any OpenAI-compatible endpoint under endpoints.custom and any Streamable HTTP MCP server under mcpServers, so one block of YAML gives every conversation Memtro's models, memory and connector tools. The key can be the organisation's, from an environment variable, or each person's own with apiKey: 'user_provided', which is what keeps personal memory and personal connections personal.

What you get

  • Custom endpoint with baseURL https://platform.memtro.com/v1: auto, memtro and every provider model you hold a key for
  • MCP server over Streamable HTTP, so memory and connector tools are in every chat
  • Per-user keys with apiKey: 'user_provided', or one organisation key from the environment
  • The model menu filled from /v1/models with models.fetch: true

How to connect

  1. 01librechat.yaml → endpoints.custom: - name: 'Memtro', apiKey: '${MEMTRO_API_KEY}' (or 'user_provided'), baseURL: 'https://platform.memtro.com/v1', models: { default: ['auto', 'memtro'], fetch: true }
  2. 02librechat.yaml → mcpServers: memtro: { type: streamable-http, url: https://platform.memtro.com/api/mcp, headers: { Authorization: 'Bearer ${MEMTRO_API_KEY}' } }
  3. 03Restart LibreChat, choose Memtro in the endpoint menu and tick its tools in the MCP menu

Example queries

  • › What changed on the Baltic account this week?
  • › auto:agent as the model runs a full investigation across your systems

Why it is better with memory

Through LibreChat you get the same memory, connector tools, routing and jobs as every other client, so what one person learns in one tool is available to the whole team in all of them.