# Valar

## Docs

- [Building a Tool-Calling Agent](https://docs.valarhq.ai/agents.md): Wire external tools into a multi-turn Valar conversation and let the model orchestrate them
- [Check batch status](https://docs.valarhq.ai/api-reference/batches-api/check-batch-status.md): See where a batch currently stands.
- [Create a batch](https://docs.valarhq.ai/api-reference/batches-api/create-a-batch.md): Submit a batch of requests for asynchronous processing.
- [Get a single batch request result](https://docs.valarhq.ai/api-reference/batches-api/get-a-single-batch-request-result.md): Pull back the result for one request in a batch, looked up by its custom_id.
- [Get batch request result](https://docs.valarhq.ai/api-reference/batches-api/get-batch-request-result.md): Retrieve the result of a specific request within a batch by its custom_id.
- [Get batch status](https://docs.valarhq.ai/api-reference/batches-api/get-batch-status.md): Retrieve the current status of a batch.
- [List batches](https://docs.valarhq.ai/api-reference/batches-api/list-batches.md): List batches with optional pagination.
- [Create a chat completion](https://docs.valarhq.ai/api-reference/chat-completions-api/create-a-chat-completion.md): OpenAI-compatible Chat Completions endpoint. Supports streaming via stream: true, which returns a Server-Sent Events stream of chat.completion.chunk objects.
- [Create an Anthropic message](https://docs.valarhq.ai/api-reference/messages-api/create-an-anthropic-message.md): Anthropic-compatible Messages endpoint. Streaming is not currently supported.
- [Create an Anthropic-style message](https://docs.valarhq.ai/api-reference/messages-api/create-an-anthropic-style-message.md): A Messages endpoint that follows the Anthropic interface. Streaming is not available here yet.
- [List supported models](https://docs.valarhq.ai/api-reference/models-api/list-supported-models.md)
- [List the models you can call](https://docs.valarhq.ai/api-reference/models-api/list-the-models-you-can-call.md)
- [Create a response](https://docs.valarhq.ai/api-reference/responses-api/create-a-response.md): Creates an OpenAI Responses API task. Returns 202 when background=true, otherwise returns 200 after completion.
- [Fetch a response](https://docs.valarhq.ai/api-reference/responses-api/fetch-a-response.md)
- [Retrieve a response](https://docs.valarhq.ai/api-reference/responses-api/retrieve-a-response.md)
- [Completion windows](https://docs.valarhq.ai/completion-windows.md): Trade turn latency for cost by choosing the completion window that fits your time budget
- [Data Processing Agreement](https://docs.valarhq.ai/dpa.md): How Valar processes, stores, and protects customer data when running inference
- [Introduction](https://docs.valarhq.ai/index.md): The inference provider tuned for high-throughput agentic work
- [Inference modes](https://docs.valarhq.ai/inference-modes.md): Pick how to run inference on Valar by how soon you need each result
- [LangChain](https://docs.valarhq.ai/integrations/langchain.md): Use Valar as the model provider in LangChain through its OpenAI-compatible integration.
- [OpenAI Agents SDK](https://docs.valarhq.ai/integrations/openai-agents.md): Run the OpenAI Agents SDK for Python against Valar instead of OpenAI
- [Vercel AI SDK](https://docs.valarhq.ai/integrations/vercel-ai.md): Use Valar with the Vercel AI SDK through the OpenAI-compatible provider
- [Models](https://docs.valarhq.ai/models.md): All models currently served by Valar
- [Pricing](https://docs.valarhq.ai/pricing.md): How Valar bills inference and the per-token rate for every model
- [Quickstart](https://docs.valarhq.ai/quickstart.md): Start using Valar with OpenAI clients
- [Sending Requests at Scale](https://docs.valarhq.ai/requests_at_scale.md): Move tens of thousands of requests through Valar without serializing them one by one
- [Structured outputs](https://docs.valarhq.ai/structured-outputs.md): Constrain a model's response to a JSON schema so you get parseable, predictable JSON across Valar's APIs
- [API Support Matrix](https://docs.valarhq.ai/support.md): What each Valar inference API accepts today, and what isn't wired up yet
- [Overview](https://docs.valarhq.ai/usage.md): Query spend, balance, and operational metrics for your account programmatically
- [Endpoints](https://docs.valarhq.ai/usage-endpoints.md): Routes for spend, balance, activity, and time-series usage metrics
- [Webhooks](https://docs.valarhq.ai/webhooks.md): Receive a callback when a background request finishes instead of polling for it

## OpenAPI Specs

- [openapi](https://docs.valarhq.ai/openapi.json)