WP HoloDev is built around a bring-your-own-API approach. You choose the AI providers and models you want to use, connect your own API credentials and remain in control of which models power your development work.
Connect OpenAI, OpenCode and OpenRouter-compatible APIs, then choose from the models available through those services. Use different models for different types of work, balance capability against cost and change your setup as new models become available.
WP HoloDev provides the development harness. You decide which AI runs inside it.
Your Models, Your API Keys, Your Choice
WP HoloDev does not require you to use one bundled AI provider or lock your development workflow to a single model.
Bring Your Own API
Connect your own supported AI accounts and use the API access you already have. Your model usage remains tied to the providers and services you choose rather than being hidden behind a separate WP HoloDev model subscription.
This gives you much more control over availability, pricing and model selection. If one provider offers a model that suits a particular project better, you can use it without changing your development environment.
Connect OpenAI
OpenAI models can be connected directly to WP HoloDev using your own API credentials.
Use OpenAI Models for Development Work
Once connected, compatible OpenAI models can be made available to the harness for coding, reasoning, debugging, project analysis and other development tasks.
You decide which available models you want to use and how they fit into your workflow. A stronger model can be reserved for demanding tasks while a faster or less expensive model handles routine development work.
Use OpenCode
WP HoloDev can also work with OpenCode as part of its flexible AI development setup.
Add More Choice to Your Coding Workflow
OpenCode gives developers another route for working with coding models and compatible AI services without forcing the entire WP HoloDev environment around one provider.
This fits the wider WP HoloDev approach. The harness manages the project, tools, permissions, testing and development workflow while the connected model provides the intelligence required for the task.
Connect Through OpenRouter
OpenRouter gives WP HoloDev access to a much wider range of AI models through one compatible API.
Choose From a Large Model Marketplace
By connecting your own OpenRouter API key, you can use supported models from different model developers without configuring a separate direct provider connection for every one.
This makes it easier to experiment with different coding and reasoning models, compare their performance and select models based on capability, speed or price.
Switch Models Without Rebuilding Your Workflow
Your project remains inside the same WP HoloDev environment when you change models. The project context, development tools, validation workflow and permissions do not need to be rebuilt simply because you want to try a different AI.
The model can change while the harness around it remains consistent.
Different Models for Different Jobs
The most powerful model is not automatically the best model for every task.
Use Efficient Models for Routine Work
Small code changes, simple investigations and straightforward development tasks can often be handled effectively by faster and less expensive models.
Using an efficient model where appropriate can reduce API costs and improve response speed without requiring the same model for every stage of a project.
Use Stronger Models When the Task Demands It
Architecture decisions, difficult debugging, security analysis and final release review can require deeper reasoning and a stronger understanding of a large codebase.
WP HoloDev allows more capable models to be used for those jobs while keeping simpler tasks on models that are better suited to everyday development work.
Intelligent Model Routing
WP HoloDev can go beyond manually selecting one model and using it for everything.
Efficient, Balanced and Expert Model Tiers
Models can be organised into different capability tiers so the harness can choose an appropriate model for the work being performed.
An efficient model can handle straightforward tasks. A balanced model can take responsibility for more involved development work. Expert models can be reserved for tasks where additional reasoning, security analysis or release confidence is worth the extra cost.
Escalate When Necessary
If a task becomes more difficult than expected or repeated validation fails, the workflow can move to a stronger model rather than repeatedly asking the same model to solve a problem it is struggling with.
This helps make model choice part of the development process rather than a setting you have to constantly manage yourself.
You Can Still Take Direct Control
Automatic routing is useful, but it should not remove developer choice.
Choose the Model Yourself
You can select a particular provider or model when you want one used for a task. This is useful when you already know which model performs well on a particular codebase or type of development work.
You can also prevent specific models from being used or change how models are assigned within a project.
Keep Cost Under Control
Bring-your-own-API gives you direct visibility over the services you are paying for.
Use Expensive Models Where They Add Value
There is little benefit in sending every minor task to the most expensive model available. WP HoloDev is designed so cheaper models can perform routine work while higher-cost models are reserved for the stages where their additional capability matters.
This makes it possible to build a model strategy around both development quality and budget rather than treating AI usage as an unlimited resource.
Track Usage Through the Harness
Model usage can be associated with projects, jobs and agents, helping you understand where tokens and API spending are being used during development.
Budgets and limits can also help prevent a complex task or agent workflow from consuming more resources than intended.
Your API Credentials Stay Protected
Connecting external AI services should not require exposing your credentials to the coding agent itself.
Keys Stay Outside the Project Context
WP HoloDev keeps provider credentials separate from ordinary prompts, project code and agent memory. The model does not need to see your API key simply because the harness is using that provider.
On Windows, supported credentials can be encrypted locally so they remain part of the controlled application environment rather than being stored as plain project configuration.
No Single-Provider Lock-In
AI models are changing quickly. A model that is the best choice today may be replaced by something faster, cheaper or more capable in the future.
The Harness Stays Consistent While Models Change
WP HoloDev separates the development environment from the model provider. Your projects, tools, permissions, testing, browser workflows, memory and validation systems belong to the harness rather than to one AI vendor.
This means you can change the intelligence behind the workflow without having to change the workflow itself.
More Providers and Model Options Are Coming
OpenAI, OpenCode and OpenRouter are the beginning of WP HoloDev's model ecosystem rather than the end of it.
Designed to Expand
The model layer is designed so additional providers and compatible services can be added as WP HoloDev develops.
That means the platform can continue to support new coding models, new provider APIs and new ways of running AI without rebuilding the development harness around each one.
WP HoloDev gives you a stable development environment while allowing the AI behind it to keep evolving.