> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orq.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Model FAQ

> Answers to common questions about models in Orq.ai: enabling models, model IDs, parameters, reasoning, and what capabilities each model supports, with links to the relevant documentation.

Frequently asked questions about models and their capabilities in **Orq.ai**. Each answer links to the page that covers the topic in full.

<AccordionGroup>
  <Accordion title="Which models are available and how do I reference them?">
    Browse the full catalog on [Supported Models](/docs/ai-gateway/supported-models). Model IDs use the `provider/` prefix, for example `openai/gpt-5.6-sol`, `anthropic/claude-sonnet-5`, or `google/gemini-3.5-flash`. Models must be enabled on the [Models page](/docs/ai-gateway/using-the-router) before they can be called.
  </Accordion>

  <Accordion title="Do I need my own provider API keys?">
    Requests route to the model provider, so provider credentials are required: connect [BYOK](/docs/ai-gateway/providers-overview) keys for direct billing and rate-limit ownership, or use **Orq.ai**-managed access via [Credits](/docs/ai-studio/organization/credits) where available. See [Billing & Usage](/docs/ai-studio/organization/billing-usage).
  </Accordion>

  <Accordion title="Can I use private or custom models?">
    Add models from private or self-hosted endpoints to the workspace model catalog. See [Private Models](/docs/ai-gateway/private-models).
  </Accordion>

  <Accordion title="How do I find a model's context window or max output tokens?">
    The [Models page](/docs/ai-gateway/using-the-router) lists the **Context Length** and **Max Output Tokens** columns for every model and can filter by context window size. The static [Supported Models](/docs/ai-gateway/supported-models) page does not list context windows or max output tokens.
  </Accordion>

  <Accordion title="Why do most chat models require a user message?">
    Claude and most chat models require the conversation to include a user message. Requests whose input is only system or developer instructions, or empty, are rejected with a `400 invalid_input` error. Tool results are grouped into user turns automatically.
  </Accordion>

  <Accordion title="How should I configure temperature and top_p?">
    Ranges are per model: Anthropic models accept `0` to `1`, OpenAI models `0` to `2`. Newer Anthropic models reject requests that set `temperature` and `top_p` together; set one or the other. Audio endpoints accept `0` to `1`. See the [Anthropic provider page](/docs/ai-studio/integrations/providers/anthropic) and [Reasoning](/docs/ai-gateway/features/reasoning).
  </Accordion>

  <Accordion title="How do reasoning models behave?">
    Each provider exposes reasoning differently: OpenAI uses `reasoning_effort`, while Google Gemini and Anthropic use a `thinking` object. When `reasoning_effort` is set, the **AI Gateway** drops `temperature` and `top_p` before forwarding the request, since those parameters are incompatible with OpenAI reasoning models. See [Reasoning](/docs/ai-gateway/features/reasoning).
  </Accordion>

  <Accordion title="What happens when I send unsupported or out-of-range parameters?">
    Model providers reject incompatible parameter combinations, for example the Anthropic error `4028` when `temperature` and `top_p` are both specified. See [Reasoning](/docs/ai-gateway/features/reasoning) for the parameters the gateway drops on reasoning models.
  </Accordion>

  <Accordion title="When does native file attachment work?">
    Native file attachment works only with models that support file input directly. For Deployments, file input is supported on OpenAI, Anthropic, and Google Gemini models. PDF support varies by model. Use the [routing rule](/docs/ai-gateway/configuration/routing-rules) "Keep file attachment requests on models that support them" to route by capability. See [Multimodal](/docs/ai-gateway/features/multimodal) and [Files](/docs/ai-gateway/features/files).
  </Accordion>

  <Accordion title="Which models support image, PDF, or audio input?">
    Image, PDF, and audio input support varies by model. Use the modality tabs on the [Models page](/docs/ai-gateway/using-the-router) to scope models by type, or check the per-endpoint sections of [Supported Models](/docs/ai-gateway/supported-models). See [Multimodal](/docs/ai-gateway/features/multimodal).
  </Accordion>

  <Accordion title="Which models support structured output?">
    Models that support structured outputs can return valid JSON matching a schema. Not all models support schemas; check the **JSON Mode** capability filter on the [Models page](/docs/ai-gateway/using-the-router) and use JSON mode as a fallback. See [Structured Outputs](/docs/ai-gateway/features/structured-outputs).
  </Accordion>

  <Accordion title="Which models support streaming?">
    All models support streaming, with per-endpoint constraints. See [Streaming](/docs/ai-gateway/features/streaming).
  </Accordion>

  <Accordion title="Which models support tool calling?">
    Tool calling is supported by most chat models, with format differences per provider. See [Tool Calling](/docs/ai-gateway/features/tool-calling).
  </Accordion>

  <Accordion title="Which models support prompt caching?">
    Prompt caching support varies by provider: Anthropic models use `cache_control` markers, Google Gemini models cache implicitly, and OpenAI models cache automatically. See [Prompt Caching](/docs/ai-gateway/features/prompt-caching).
  </Accordion>

  <Accordion title="Which model should I pick, or can Orq.ai choose per request?">
    The [Smart Router](/docs/ai-gateway/smart-router) selects the optimal model from a pool per request based on task complexity and the chosen mode. For manual selection, compare models on the [Models page](/docs/ai-gateway/using-the-router).
  </Accordion>

  <Accordion title="If a provider fails or rate-limits, can traffic fail over?">
    Configure [fallbacks and retries](/docs/ai-gateway/features/retries) or [load balancing](/docs/ai-gateway/features/load-balancing) so requests flow to another model automatically when the primary is unavailable.
  </Accordion>
</AccordionGroup>