# Valkyrie

> Fine-tune, deploy, and serve LLMs on on-demand cloud GPUs, with an OpenAI-compatible API and a built-in MCP server.

- [Welcome to Valkyrie](https://docs.valkyrieapp.azumo.com/index.md): Run, fine-tune, and serve open LLMs on rented GPUs, without managing any infrastructure.
- [Quickstart](https://docs.valkyrieapp.azumo.com/quickstart.md): From sign-up to your first OpenAI-compatible API call.
- [Core concepts](https://docs.valkyrieapp.azumo.com/concepts.md): The building blocks of Valkyrie: deployments, fine-tuning jobs, GPUs, the wallet, aliases, and the MCP server.
- [Datasets](https://docs.valkyrieapp.azumo.com/guides/datasets.md): Upload and manage the datasets you fine-tune on.
- [Fine-tune a model](https://docs.valkyrieapp.azumo.com/guides/fine-tuning.md): Adapt an open base model to your data with a managed Unsloth pipeline.
- [Deploy a model](https://docs.valkyrieapp.azumo.com/guides/deploy-model.md): Serve a fine-tuned model or any Hugging Face model behind an OpenAI-compatible endpoint.
- [Run inference](https://docs.valkyrieapp.azumo.com/guides/inference.md): Call your deployed models with the OpenAI-compatible Chat Completions API.
- [Aliases](https://docs.valkyrieapp.azumo.com/guides/aliases.md): Give a deployment a stable, human-readable URL.
- [Quota & billing](https://docs.valkyrieapp.azumo.com/guides/wallet-billing.md): How Valkyrie's prepaid balance and per-use pricing work.
- [Connect a model](https://docs.valkyrieapp.azumo.com/connect/overview.md): Use a Valkyrie-hosted model as the LLM inside your tools, agents, and SDKs.
- [Claude Code](https://docs.valkyrieapp.azumo.com/connect/claude-code.md): Use a Valkyrie-hosted model as Claude Code's model.
- [Cursor](https://docs.valkyrieapp.azumo.com/connect/cursor.md): Add a Valkyrie model to Cursor as a custom OpenAI-compatible model.
- [OpenAI-compatible clients](https://docs.valkyrieapp.azumo.com/connect/openai-compatible.md): The generic recipe for connecting any OpenAI-compatible SDK or app to Valkyrie.
- [Dashboard overview](https://docs.valkyrieapp.azumo.com/dashboard/overview.md): A tour of the Valkyrie web dashboard and its screens.
- [API Keys](https://docs.valkyrieapp.azumo.com/dashboard/api-keys.md): Create and manage the keys applications and agents use to call Valkyrie.
- [Models](https://docs.valkyrieapp.azumo.com/dashboard/models.md): Deploy, operate, and share your models from the dashboard.
- [Fine-Tuning screen](https://docs.valkyrieapp.azumo.com/dashboard/fine-tuning.md): Create and monitor training jobs from the dashboard.
- [Usage](https://docs.valkyrieapp.azumo.com/dashboard/usage.md): Per-request telemetry for calls to your deployed models.
- [Quota](https://docs.valkyrieapp.azumo.com/dashboard/quota.md): Your prepaid balance and every charge Valkyrie has billed against it.
- [Inference API](https://docs.valkyrieapp.azumo.com/api-reference/introduction.md): Call your deployed Valkyrie models over an OpenAI- or Anthropic-compatible API.
- [Authentication](https://docs.valkyrieapp.azumo.com/api-reference/authentication.md): How to authenticate requests to the Valkyrie API.
- [Chat Completions](https://docs.valkyrieapp.azumo.com/api-reference/chat-completions.md): The OpenAI-compatible chat completions endpoint.
- [Errors](https://docs.valkyrieapp.azumo.com/api-reference/errors.md): Common HTTP status codes and how to handle them.
- [MCP server overview](https://docs.valkyrieapp.azumo.com/mcp/overview.md): Operate Valkyrie from an AI agent through the Model Context Protocol, what it is, and what it can and can't do.
- [Set up the MCP server](https://docs.valkyrieapp.azumo.com/mcp/setup.md): Register the Valkyrie MCP server in your client, with the transport and auth details.
- [MCP tools reference](https://docs.valkyrieapp.azumo.com/mcp/tools.md): Every tool the Valkyrie MCP server exposes, grouped by area, with what each one does.