MindsDB

Integrate MindsDB with Portkey for production-grade AI pipelines

MindsDB connects data sources and LLMs for AI automation. Portkey adds:

• 1600+ LLMs — Access any provider through MindsDB
• Observability — Track costs, latency, and usage
• Reliability — Caching, routing, guardrails via Configs

Prerequisites

1. Install MindsDB via Docker or Docker Desktop
2. Install Portkey dependencies per MindsDB instructions
3. Get your Portkey API key

Setup

1. Create AI Engine

   Using Config ID

   CREATE ML_ENGINE portkey_engine
   FROM portkey
   USING
       portkey_api_key = 'PORTKEY_API_KEY',
       config = 'pc-your-config-id';
   

   CREATE ML_ENGINE portkey_engine
   FROM portkey
   USING
       portkey_api_key = 'PORTKEY_API_KEY',
       provider = '@openai-prod';
   

   Use Configs to add guardrails, caching, conditional routing, and more.

2. Create Model

   CREATE MODEL portkey_model
   PREDICT answer
   USING
       engine = 'portkey_engine',
       temperature = 0.2;
   
3. Query

   SELECT question, answer
   FROM portkey_model
   WHERE question = 'Where is Stockholm located?';
   

   Output:

   +-----------------------------+--------------------------------------------------------------------------------+
   | question                    | answer                                                                         |
   +-----------------------------+--------------------------------------------------------------------------------+
   | Where is Stockholm located? | Stockholm is the capital of Sweden, located on the south-central east coast... |
   +-----------------------------+--------------------------------------------------------------------------------+
   

Next Steps

MindsDB Use Cases

Explore MindsDB examples

Portkey Configs

Add routing, caching, guardrails