# Morse API docs > One OpenAI-compatible API for 80+ open-weight models, called from Morse's attested enclave, which keeps no prompt or reply. ## Docs: Get started - [Overview](https://trymorse.ai/inference/docs): One OpenAI-compatible API for 80+ open-weight models, called from Morse's attested enclave, which keeps no prompt or reply. - [Quickstart](https://trymorse.ai/inference/docs/quickstart): Make a key, then send your first request with curl, Python, or TypeScript. - [Authentication](https://trymorse.ai/inference/docs/authentication): Send your key as a Bearer token, keep it on a server, and limit what each key can do. - [Switch from OpenAI or OpenRouter](https://trymorse.ai/inference/docs/migrate): Change the base URL, the key, and the model id, and see the few fields Morse treats differently. - [FAQ](https://trymorse.ai/inference/docs/faq): Short answers about privacy, prices, limits, and what the API supports. ## Docs: Privacy and models - [Models](https://trymorse.ai/inference/docs/models): Model ids, how to list them, and what each model can do on each endpoint. - [Privacy tiers](https://trymorse.ai/inference/docs/privacy-tiers): What Anonymous, Private, TEE, and E2EE each protect, and what they don't. - [Provider routing](https://trymorse.ai/inference/docs/provider-routing): Set a privacy floor for a request, and choose which providers may serve it. - [What Morse keeps](https://trymorse.ai/inference/docs/data-retention): What Morse records about a request, what it never records, and what a provider sees. - [Verification and attestation](https://trymorse.ai/inference/docs/verification): Check that Morse's enclave runs the published release, and read the receipt on a response. - [Sealed API keys](https://trymorse.ai/inference/docs/sealed-api-keys): Seal each request on your machine, so Morse's relay sees only ciphertext, with morse-seal or attested(). ## Docs: Features - [Streaming](https://trymorse.ai/inference/docs/streaming): Receive a reply as the model writes it, get the token counts, and handle an error mid-stream. - [Tool calling](https://trymorse.ai/inference/docs/tool-calling): Let a model call your functions, with OpenAI's tools and tool_choice. - [Reasoning](https://trymorse.ai/inference/docs/reasoning): Choose how hard a reasoning model thinks with reasoning_effort. - [Images as input](https://trymorse.ai/inference/docs/vision): Send images in a message as data URIs. - [Image generation](https://trymorse.ai/inference/docs/image-generation): Generate up to four images from a prompt and reference images. - [Web search and page reading](https://trymorse.ai/inference/docs/web-search): Search the web and read a page, for agents that need fresh facts. - [Audio transcription](https://trymorse.ai/inference/docs/transcription): Turn a recording of speech into text, in an enclave. - [Prompt caching](https://trymorse.ai/inference/docs/prompt-caching): How cached input is priced, and how Morse keeps a conversation on a warm cache. - [End-to-end encryption](https://trymorse.ai/inference/docs/e2ee): Encrypt each request to a key only the model's enclave holds. - [Privacy Veil](https://trymorse.ai/inference/docs/privacy-veil): Swap names, addresses, and other personal details for placeholders before a provider sees them. - [Presets](https://trymorse.ai/inference/docs/presets): Save a system prompt and examples once, and refer to them by id. ## Docs: Frameworks and SDKs - [OpenAI SDK](https://trymorse.ai/inference/docs/openai-sdk): Use the official OpenAI SDKs for Python and TypeScript with Morse. - [Vercel AI SDK](https://trymorse.ai/inference/docs/vercel-ai-sdk): Call Morse from the AI SDK with its OpenAI-compatible provider. - [LangChain](https://trymorse.ai/inference/docs/langchain): Use Morse as a chat model in LangChain. - [LiteLLM](https://trymorse.ai/inference/docs/litellm): Add Morse to a LiteLLM proxy or SDK call. - [Coding agents and editors](https://trymorse.ai/inference/docs/coding-agents): Point Cline, OpenCode, Zed, Aider, Goose, and other tools at Morse. ## Docs: Account - [Pricing and billing](https://trymorse.ai/inference/docs/pricing): How requests are priced, the order credits are spent in, and what a hold is. - [Rate limits](https://trymorse.ai/inference/docs/rate-limits): The requests a minute and in flight each plan allows, and how to retry a 429. - [Errors](https://trymorse.ai/inference/docs/errors): The error shape every failed request returns, the codes at each status, and when to retry. - [Response headers](https://trymorse.ai/inference/docs/response-headers): The headers on every reply, and what each tells you about the request. ## API reference: Introduction - [API reference](https://trymorse.ai/inference/docs/api): The base URL, authentication, the request and reply as types, finish reasons, and cost. - [Streaming format](https://trymorse.ai/inference/docs/api/streaming): The server-sent events a streamed reply sends, and how a stream ends. - [Chat parameters](https://trymorse.ai/inference/docs/api/parameters): Which chat parameters reach the model, which Morse checks, which it refuses, and which it drops. - [Limits](https://trymorse.ai/inference/docs/api/limits): Body sizes, message counts, media, rate limits, and credit, for every endpoint. - [Error codes](https://trymorse.ai/inference/docs/api/errors): Every error code the API returns, by status, and when to retry. ## API reference: Endpoints - [Create a chat completion](https://trymorse.ai/inference/docs/api/chat-completions): Send a conversation to a model and get its reply, whole or streamed. - [List models](https://trymorse.ai/inference/docs/api/models): Every model, with its type and the tiers it's sold at, each with its limits and price. - [Retrieve a model](https://trymorse.ai/inference/docs/api/model): One model, by its id or alias, as the model list shows it. - [Generate images](https://trymorse.ai/inference/docs/api/images): Generate up to four images from a prompt and reference images. - [Search the web](https://trymorse.ai/inference/docs/api/search): Run up to three web searches and get the top pages. - [Read a web page](https://trymorse.ai/inference/docs/api/fetch): Read one public web page and get its text. - [Transcribe audio](https://trymorse.ai/inference/docs/api/transcriptions): Turn a recording of speech into text. - [Get the version](https://trymorse.ai/inference/docs/api/version): The build the API is running. No key needed. ## OpenAPI - [OpenAPI spec](https://trymorse.ai/inference/docs/openapi.json): the whole API as an OpenAPI 3.1 file