# Modular Cloud

## Docs

- [API keys](https://docs-preview.modular.com/administration/api-keys.md): Create and manage API keys for authenticating your requests
- [Billing](https://docs-preview.modular.com/administration/billing.md): Billing tiers, endpoint access, and where to manage payment
- [Manage users](https://docs-preview.modular.com/administration/manage-users.md): Invite team members and manage their roles in your organization
- [Create deployment](https://docs-preview.modular.com/api-reference/deployments/create-deployment.md): Create a new dedicated inference endpoint. `name` and `model_id` are immutable after creation.
- [Get deployment](https://docs-preview.modular.com/api-reference/deployments/get-deployment.md): Get details for a single deployment, including current status and endpoint URL.
- [List deployments](https://docs-preview.modular.com/api-reference/deployments/list-deployments.md): List all deployments in the organization.
- [Terminate deployment](https://docs-preview.modular.com/api-reference/deployments/terminate-deployment.md): Soft-delete a deployment. The deployment enters `terminating` status and is cleaned up asynchronously.
- [Update deployment](https://docs-preview.modular.com/api-reference/deployments/update-deployment.md): Update mutable fields on a deployment. Uses deep-merge semantics on `configuration` so omitted fields are left unchanged. `name` and `model_id` cannot be changed.
- [Create chat completion](https://docs-preview.modular.com/api-reference/inference/create-chat-completion.md): Generate a response from a conversation. Accepts text, image, and video input for multimodal models, and supports function calling via `tools`/`tool_choice`. Request and response shapes match the OpenAI Chat Completions API, so existing OpenAI SDK clients work unchanged.
- [Create response](https://docs-preview.modular.com/api-reference/inference/create-response.md): Generate an image or video from a text prompt, or transform an existing image. Built on the [Open Responses](https://huggingface.co/blog/open-responses) spec, a provider-agnostic API standard.
- [Create a deployment](https://docs-preview.modular.com/deployments/create.md): Create a dedicated inference endpoint via the console or API
- [Function calling and tool use](https://docs-preview.modular.com/inference/function-calling.md): Implement OpenAI-compatible function calling and tool use with the hosted inference API
- [Image generation](https://docs-preview.modular.com/inference/image.md): Generate and transform images using image generation models via the v1/responses endpoint
- [Text generation](https://docs-preview.modular.com/inference/text.md): Generate text and analyze images and video using an OpenAI-compatible chat completions endpoint
- [Video generation](https://docs-preview.modular.com/inference/video.md): Generate videos from text prompts using video generation models via the v1/responses endpoint
- [Supported models](https://docs-preview.modular.com/models.md): Find the right model for your use case, and see which endpoints, modalities, and features it supports
- [Quickstart](https://docs-preview.modular.com/quickstart.md): Make your first inference request in minutes

## OpenAPI Specs

- [openapi](https://docs-preview.modular.com/reference/openapi.json)
- [openapi-inference](https://docs-preview.modular.com/reference/openapi-inference.json)
- [openapi-deployments](https://docs-preview.modular.com/reference/openapi-deployments.json)