EidoStack / Documentation
Learn to explore and test AI models in EidoStack.
From your first model test to more advanced experiments, this documentation helps you get the most out of EidoStack. Learn how to connect your own AI provider accounts, try supported models, experiment with prompts and context, compare responses, and understand token usage and estimated costs in one browser-based workspace. Start with the essentials, then explore the tools and settings you need for hands-on model testing. Available input types and other capabilities depend on the selected model and what the Playground currently supports.
Explore the PlaygroundExplore the workspace
Documentation sections
Getting Started
Playground and Model Testing
Playground workspace
Model experiments
Models and Providers
Settings
Model behavior
Usage and Costs
Account and Billing
Account
Subscription
Security and Data
Vault recovery
Support
Common Questions
What can I do in the EidoStack Playground?
EidoStack lets developers test and explore supported AI models in a single browser-based workspace. You can run your own prompts, experiment with system instructions and conversation context, compare model responses, and inspect token usage and estimated API costs. It is designed for hands-on experimentation rather than automated benchmarks or model training.
How do I get started with EidoStack?
Create an account, set up your workspace, connect an AI provider using your own API key, and start your first chat. The Getting Started guides walk you through each step so you can begin testing models without installing additional software.
How do I connect AI models from different providers?
EidoStack supports models from providers such as OpenAI, Anthropic, and Google AI. Add your own provider credentials through the API Key Vault, then select a supported model in your workspace. The provider and model guides explain setup and model selection. Your provider account is responsible for API usage charges.
Can I compare model responses in the Playground?
Yes. Use the comparison workflow to send the same prompt to supported models and inspect their responses side by side. Compare output quality and behavior alongside available token and estimated cost information. This helps you decide which model fits your own task rather than relying only on general rankings.
How does EidoStack handle multi-model capabilities?
Different AI models support different kinds of input and output. EidoStack's multi-model Playground is built around those model-specific capabilities rather than assuming every model works the same way. The inputs and controls you can use depend on the selected model, provider, and capabilities currently available in EidoStack. Check the model documentation and selector for supported options.
Where can I see token usage, context, and API costs?
The Usage and Costs guides explain how to inspect token consumption and estimated provider costs. The context documentation explains how system prompts and conversation history affect what a model receives. Use these measurements to understand the trade-offs while testing, not just after you choose a model.
How are my AI provider API keys protected?
For normal model testing, EidoStack stores provider credentials in an encrypted vault in your browser rather than in the EidoStack application database. The Security and Data guides explain how the vault works, how credentials are used for provider requests, and how to create and restore an encrypted backup. Your prompts and selected context are still sent to the AI provider you choose.