The Synthesis Graph™ is not directly accessible via API. There is no graph query endpoint — the knowledge graph is queried internally during model inference. To retrieve knowledge from the graph, call a model inference endpoint; the model traverses the graph as part of generating its response.
Accessing 4MINDS API
The API is a dedicated tab in the main navigation. It’s available on the Enterprise plan and only visible to users with Admin rights — if you don’t see the tab, confirm your plan and role with your organization administrator.
From the API tab, you can view your API keys, configure access, and browse the API reference.
The ‘Quick Start’ section in the left sidebar lets you select the API version and a resource from the dropdowns to view all available 4MINDS API endpoints.
4MINDS offers two API versions, selectable from the first dropdown in the Quick Start section: the regular 4MINDS API and the OpenAI-compatible API. Use the OpenAI-compatible version if you’re migrating from OpenAI or want to reuse existing OpenAI SDK code with minimal changes.
The table below the ‘Quick Start’ section contains resource IDs required for certain API requests. Available resources in the dropdown include: models, base models, datasets, training types, use cases, fine-tuning categories and personas.
Authentication
To start working with the 4MINDS API, click the ‘Create API Key’ button to generate your authentication key.
Configure your API key by entering a name, selecting the key type (Universal or Model-specific), setting permissions (All Access, Read Only, Inference Only, or Custom) and expiration period. Click the ‘Create Key’ button to finalize and generate your authentication key.
You’ll receive a confirmation that your key has been created successfully.
Copy your API authentication key now and store it securely, as you won’t be able to retrieve it later.
Base URL
Support
For API access or questions, contact support at support@4minds.ai.
