AI INFERENCE

From an idea to an API call.

Connect your application to language models through one consistent API. Keep access, usage and service credit in the same workspace.

Open your console ↗Read the documentation

Platform preview · Accounts and project requests are open. AI inference and card funding are awaiting activation.

{ }

A familiar developer interface

Use a JSON chat-completions interface for assistants, document workflows and application features. Select an activated model, send your messages and receive an answer with token usage and a request reference.

Bearer API-key authentication
Text-based chat completions
Per-model input and output pricing
Usage returned with each successful request

↗

Usage you can explain

Every completed request has its own usage record. See the model, input and output tokens, service-credit charge and response time. Set a monthly spending limit when you create a key.

Account-level prepaid service credit
Separate keys for separate applications
Atomic credit reservation before inference
Revocation without changing your account password

PRACTICAL APPLICATIONS

Build features people use.

Product assistants

Add conversational interactions to your product. Keep application instructions in your server and use a dedicated API key with a controlled budget.

Document workflows

Send extracted text for summarisation, classification or drafting. Choose a suitable model and review your own data-handling requirements before sending sensitive information.

Internal tools

Connect inference to your application backend, workflows and internal software. Use usage records to allocate costs across your own projects.

Model availability is shown in the live catalog and console. RDGlobal does not guarantee that every model is available in every location or for every workload.