> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.merge.dev/merge-gateway/capabilities/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.merge.dev/_mcp/server.

# Request capabilities

> Build once. Connect to every API, tool, and LLM. Ship products and agents to production without owning the integration stack.

## Docs

- [Multimodal](https://docs.merge.dev/merge-gateway/capabilities/multimodal/overview.md): Which endpoint and SDK surface handles each modality in Gateway, the inline size caps for media in chat requests, and how to find models that support each one.
- [Images](https://docs.merge.dev/merge-gateway/capabilities/multimodal/images.md): Send images to vision-capable models on Gateway with the image_url or image content block, the OpenAI SDK, or AI SDK file parts.
- [Documents](https://docs.merge.dev/merge-gateway/capabilities/multimodal/documents.md): Send PDF documents to document-capable models on Gateway as base64 or a URL with the document content block or AI SDK file parts.
- [Image generation](https://docs.merge.dev/merge-gateway/capabilities/multimodal/image-generation.md): Generate images on Gateway with POST /v1/images/generations or inline on /v1/responses with image-capable chat models, and read per-token or per-image pricing.
- [Image editing](https://docs.merge.dev/merge-gateway/capabilities/multimodal/image-editing.md): Edit images on Gateway with POST /v1/images/edits: the multipart request, reference images and masks, upload limits, errors, and image-input billing.
- [Text to speech](https://docs.merge.dev/merge-gateway/capabilities/multimodal/text-to-speech.md): Generate speech on Gateway with POST /v1/audio/speech: raw audio bytes in the response, request fields, vendor-specific voices, and per-character billing.
- [Speech to text](https://docs.merge.dev/merge-gateway/capabilities/multimodal/speech-to-text.md): Transcribe audio on Gateway with POST /v1/audio/transcriptions: the multipart request, the JSON transcript response, duration-based billing, and size limits.
- [Live voice](https://docs.merge.dev/merge-gateway/capabilities/multimodal/live-voice.md): Connect to OpenAI GPT-Live (openai/gpt-live-1) through Gateway at wss://api-gateway.merge.dev/v1/live/sessions: authentication, session.start, delegation, per-second billing, limits, and close codes.
- [Video generation](https://docs.merge.dev/merge-gateway/capabilities/multimodal/video-generation.md): Generate video on Gateway through the async POST /v1/videos job flow: submit a prompt with optional media, poll status, download the MP4, and read per-second billing.
- [Multi-turn conversations](https://docs.merge.dev/merge-gateway/capabilities/multi-turn-conversations.md): Run multi-turn conversations on Gateway by sending the full history in input, or by storing a turn with store and resuming it with previous_response_id.
- [Tool calling](https://docs.merge.dev/merge-gateway/capabilities/tool-calling.md): Define tools, send tool-calling requests, handle tool_use blocks, return tool results, and control tool_choice on Gateway across the native API, OpenAI SDK, and AI SDK.
- [Structured outputs](https://docs.merge.dev/merge-gateway/capabilities/structured-outputs.md): Request JSON output on Gateway with response_format, control strict mode, find capable routes, and understand emulation, failover, and error behavior.
- [Decisions](https://docs.merge.dev/merge-gateway/capabilities/decisions.md): Call decision models on Gateway with POST /v1/decisions: define noul, choice, and score questions about a state, read typed answers with probabilities, and know the limits.
- [Reasoning](https://docs.merge.dev/merge-gateway/capabilities/reasoning.md): Request reasoning with an effort level or a thinking budget, check which routes support it, replay signed thinking blocks in tool loops, and read reasoning token usage.
- [Prompt caching](https://docs.merge.dev/merge-gateway/capabilities/prompt-caching.md): How Gateway handles cache markers on explicit routes and session hints on automatic routes, and how to read cache token usage on each surface.
- [Service tiers](https://docs.merge.dev/merge-gateway/capabilities/service-tiers.md): Request the flex, priority, or ultrafast service tier, opt into standard-tier fallback when the vendor throttles, and read the tier you were billed at.
- [Batch inference](https://docs.merge.dev/merge-gateway/capabilities/batch-inference.md): Submit, track, and collect batches on Gateway across chat, embeddings, Messages, and Responses, plus how batch lines are billed and the limits and errors to plan for.
- [Web search](https://docs.merge.dev/merge-gateway/capabilities/web-search.md): Add Gateway-hosted web search to a request, choose a search engine, and read citations, usage, and per-engine pricing.
- [Fusion](https://docs.merge.dev/merge-gateway/capabilities/fusion.md): Run a panel of models on the same prompt with Gateway Fusion, then have a judge model synthesize one answer.
- [Bring your own keys (BYOK)](https://docs.merge.dev/merge-gateway/capabilities/byok/overview.md): Connect your own AI vendor accounts to Gateway so requests are billed to your vendor account. Covers adding credentials, managed fallback, and which credentials requests use.
- [Amazon Bedrock](https://docs.merge.dev/merge-gateway/capabilities/byok/amazon-bedrock.md): Set up Amazon Bedrock BYOK on Gateway with a Bedrock API key, AWS access keys, or an IAM role, including the AWS-side steps and the IAM permissions each method needs.