Agentsql vs mnemiq

AgentSQL is a web service that turns plain-English questions about connected databases into read-only SQL queries, then returns a chart, table, and plain-English answer. Each response shows the exact SQL, and users can rephrase their question to have the service rewrite and rerun it. Supported connections are Postgres, MySQL, Snowflake, and BigQuery. Connections use least-privilege read-only credentials, can be limited to selected tables, and can be revoked. AgentSQL says it encrypts connections in transit, does not copy database tables or build a shadow warehouse, and does not use database contents, queries, or results to train AI models. It stores query history; Team and Scale also retain result rows. Plans are billed yearly: Starter costs $24 per month, Team $59 per month, Scale $179 per month, and Enterprise $299 per month. Monthly allowances are 600, 1,500, 5,000, and 8,000 questions respectively, with no more than one tenth usable in a day. The service targets founders, operations teams, analysts, and product managers.

mnemiq is an open-source text-to-SQL engine for enterprise data agents, configured and measured against a user's own database. Before execution, a deterministic check verifies that proposed SQL is read-only, uses permitted objects, compiles in the source dialect, and passes EXPLAIN. Access permissions apply before schema retrieval, so the model is not shown tables the caller cannot access. Each answer includes the SQL run, the tables it touched, and the enrichment version used. mnemiq can profile columns and build descriptions, grain, glossary terms, and coded-value meanings; a human owner can certify semantic definitions. Listed databases are PostgreSQL, SQLite, DuckDB, Oracle, Snowflake, and Databricks, with DuckDB as the universal executor. Models can use any OpenAI-compatible /v1 endpoint, including local servers such as vLLM, Ollama, llama.cpp server, and LM Studio. Its MCP server offers two read-only, access-scoped tools: db_read(question) and get_schema(). One process can also serve a workbench and HTTP endpoints for asking, chatting, and retrieving schema. The Apache-2.0 engine is free and runs in the user's environment. Local model hosting keeps schemas, questions, and rows inside the user's network, according to the maker.