Multi-Provider AI Support
What It Does
Multi-Provider AI Support gives you the freedom to choose the AI engine that works best for your team. An administrator sets it once in Settings → AI Settings for the whole organization: pick a model, enter the key, and use Test Connection to check it. Refresh available models asks the provider which models your key can use, so new releases appear without waiting for us. For teams that cannot send specs or prompts outside their network, Trial and Enterprise can point every AI feature at a self-hosted, OpenAI-compatible server, and an installation can refuse AI-assisted generation over MCP unless the model is self-hosted. Embeddings are picked automatically from the keys you have, ending in a built-in fallback that needs no key at all. AI stays optional throughout: test generation, gap filling and failure triage run deterministically without it, while the Project Assistant and Fix with AI need a configured model.
Overview
Not locked into one AI vendor. Bring your own key from 19 provider families, including OpenAI (GPT-5.x, GPT-4.1, GPT-4o), Anthropic Claude, Google Gemini, Azure OpenAI, Mistral, Groq, DeepSeek and OpenRouter, or run on your own model through any OpenAI-compatible server such as Ollama, LM Studio or vLLM on Trial and Enterprise. Your own keys give you full cost control; they are encrypted at rest and checked before every generation. Embeddings follow the key you configure (OpenAI, then Gemini, Cohere, a self-hosted model or a built-in fallback that needs no key). Switch providers at any time without losing tests, coverage data or execution history. And with no AI key at all, test generation, gap filling and failure triage still work.
Key Capabilities
How It Works
- 1
An administrator opens Settings → AI Settings and picks a provider and model
- 2
Enter your API key and click Test Connection
- 3
For a self-hosted model on Trial or Enterprise, set the base URL and refresh the model list
- 4
Save; every AI feature in the organization uses the selected model
- 5
Switch providers anytime without affecting existing tests or data
Available on
All Plans
Included in the free trial — no credit card required.
Documentation
Step-by-step guides for setting up and using Multi-AI.
Built for regulated teams
Teams in regulated industries rely on this capability to ship API changes with an audit trail.
Go deeper
Guides and playbooks that show this capability in practice.
Related Features
AI Test Generation
Generate a structured test suite from OpenAPI, Swagger, WSDL, GraphQL or Postman with one click, with or without an AI key.
Learn moreContinuous Learning
Every passing run, every edit and every requirement document teaches the platform how your API really behaves.
Learn moreWorkflows
Build multi-step API flows with a visual DAG editor.
Learn moreTry Multi-AI Today
Start your 15-day free trial — no credit card required.