# MindsHub Inference > Call every major model through the OpenAI, Responses and Anthropic APIs, with one key and one bill. MindsHub Inference is an OpenAI compatible inference API. Every page below is also available as markdown at the same address with a .md suffix. llms-full.txt holds the whole documentation set in one file. ## Table of Contents - [Cancel a background response](https://docs.mindshub.ai/inference/api-reference/cancel-response.md): Only a response created with `background: true` can be cancelled, as on OpenAI. Every request completes synchronously, so a stored background respo... - [chat](https://docs.mindshub.ai/inference/api-reference/chat.md): chat - [Count tokens](https://docs.mindshub.ai/inference/api-reference/count-message-tokens.md): Prompt tokens a Messages request would consume. Exact for a concrete Claude model name; an estimate from the platform's Claude counting model for a... - [Create a chat completion](https://docs.mindshub.ai/inference/api-reference/create-chat-completion.md): OpenAI Chat Completions, on any model in the catalog. Send `model` (an alias) and `messages`; set `stream: true` for `text/event-stream` chunks ter... - [Evaluate typed decisions](https://docs.mindshub.ai/inference/api-reference/create-decisions.md): Evaluate named noul, choice, or score questions against shared state. - [Create embeddings](https://docs.mindshub.ai/inference/api-reference/create-embedding.md): OpenAI embeddings on an embedding alias (`embed-small`). Always bills to the wallet, never to the included allowance. The response's `model` names ... - [Create a message](https://docs.mindshub.ai/inference/api-reference/create-message.md): Anthropic Messages, on any model in the catalog. Authenticate with `Authorization: Bearer` (the SDK's `auth_token`), never `x-api-key`. `model` tak... - [Create a response](https://docs.mindshub.ai/inference/api-reference/create-response.md): OpenAI Responses, on any model in the catalog. Send `model` and `input`; set `stream: true` for the typed `response.*` event stream. Turns are stor... - [decisions](https://docs.mindshub.ai/inference/api-reference/decisions.md): decisions - [Delete a stored response](https://docs.mindshub.ai/inference/api-reference/delete-response.md): Hard delete: the content is gone, not flagged. Deleting an absent id is a 404, as on OpenAI. - [embeddings](https://docs.mindshub.ai/inference/api-reference/embeddings.md): embeddings - [Retrieve a stored response](https://docs.mindshub.ai/inference/api-reference/get-response.md): The response as it was returned, output items included. Stored turns expire after 30 days. - [Get a trace](https://docs.mindshub.ai/inference/api-reference/get-trace.md): One trace with its span waterfall and payloads. Content is visible only to the member who made the request. 403 when you can see the organization's... - [List models](https://docs.mindshub.ai/inference/api-reference/list-models.md): Every alias in the catalog with the metadata a model picker needs, including whether your organization can call it right now. Free to call, never p... - [List a stored response's input items](https://docs.mindshub.ai/inference/api-reference/list-response-input-items.md): That turn's own input items, not the inherited history. Paginated with `after`. - [List traces](https://docs.mindshub.ai/inference/api-reference/list-traces.md): Per-request records for your organization, newest first. Members see their own requests; admins see the organization's. Available where the organiz... - [messages](https://docs.mindshub.ai/inference/api-reference/messages.md): messages - [MindsHub Inference API](https://docs.mindshub.ai/inference/api-reference/mindshub-inference-api.md): Every major model through the three request formats you already use: OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages, on one key ... - [models](https://docs.mindshub.ai/inference/api-reference/models.md): models - [responses](https://docs.mindshub.ai/inference/api-reference/responses.md): responses - [traces](https://docs.mindshub.ai/inference/api-reference/traces.md): traces - [Authentication](https://docs.mindshub.ai/inference/authentication.md): API keys: creating, using, and revoking them. - [Billing](https://docs.mindshub.ai/inference/billing.md): The included allowance, prepaid wallet, metering, and price list. - [Bring your own keys](https://docs.mindshub.ai/inference/byok.md): Planned, not yet available: run inference on your own provider API keys. - [Capability matrix](https://docs.mindshub.ai/inference/capability-matrix.md): What each API and each model supports, derived from the same code and policy the router runs. - [Chat Completions](https://docs.mindshub.ai/inference/chat-completions.md): POST /v1/chat/completions: the OpenAI chat completions shape, with any MindsHub model behind it. - [Choosing an API](https://docs.mindshub.ai/inference/choosing-an-api.md): Choose chat, embeddings, or decisions, then select the request format your application needs. - [MindsHub in Codex](https://docs.mindshub.ai/inference/codex.md): Run OpenAI Codex on any MindsHub model, Claude and Kimi included, with one key. - [Coding agents](https://docs.mindshub.ai/inference/coding-agents.md): Run Claude Code and OpenAI Codex on any MindsHub model, billed to one balance. - [Core concepts](https://docs.mindshub.ai/inference/concepts.md): The handful of ideas the rest of the documentation assumes. - [Conversation state](https://docs.mindshub.ai/inference/conversation-state.md): Where conversation history lives: client-side on two APIs, server-side on Responses. - [Decision models](https://docs.mindshub.ai/inference/decision-models.md): Understand Jev, choose a question type, and turn model judgments into predictable application behavior. - [Decisions (Jev)](https://docs.mindshub.ai/inference/decisions.md): POST /v1/decisions: runnable examples, typed questions and answers, model aliases, usage, limits, and errors. - [Embeddings](https://docs.mindshub.ai/inference/embeddings.md): POST /v1/embeddings: the OpenAI embeddings shape. - [Errors](https://docs.mindshub.ai/inference/errors.md): Status codes, body shapes, and retry guidance for every error the API returns. - [Getting started](https://docs.mindshub.ai/inference/getting-started.md): One key, every chat model, through the three chat APIs you already use. - [Images](https://docs.mindshub.ai/inference/images.md): Send images to vision-capable models as data URLs or public URLs. - [MindsHub Inference](https://docs.mindshub.ai/inference/index.md): Chat, embeddings, and Jev decisions through MindsHub Inference, with one key and one bill. - [Kimi K3 in Claude Code](https://docs.mindshub.ai/inference/kimi-claude-code.md): Run Claude Code on Kimi K3 with a MindsHub API key. - [Messages](https://docs.mindshub.ai/inference/messages.md): POST /v1/messages: the Anthropic Messages shape, with any MindsHub model behind it. - [Models endpoint](https://docs.mindshub.ai/inference/models-endpoint.md): List the catalog your key can see, with per-model metadata. - [Models](https://docs.mindshub.ai/inference/models.md): The model catalog, aliases, and reasoning effort levels. - [Prompt caching](https://docs.mindshub.ai/inference/prompt-caching.md): Repeated prompt prefixes bill at a fraction of the input rate, automatically. - [Rate limits](https://docs.mindshub.ai/inference/rate-limits.md): Throughput limits: requests, tokens, and concurrency per minute. - [Reasoning](https://docs.mindshub.ai/inference/reasoning.md): Control how hard a model thinks before it answers, and what that costs. - [Responses](https://docs.mindshub.ai/inference/responses.md): POST /v1/responses: the OpenAI Responses shape, with any MindsHub model behind it. - [Streaming](https://docs.mindshub.ai/inference/streaming.md): Stream tokens as they are generated, in each API's native event format. - [Structured output](https://docs.mindshub.ai/inference/structured-output.md): Ask for a JSON schema and get JSON that matches it, on any model through any API. - [Tool calling](https://docs.mindshub.ai/inference/tool-calling.md): Declare functions, receive calls, return results, on any model through any API. - [Traces](https://docs.mindshub.ai/inference/traces.md): Read your organization's request history, and drill into one request's spans and payloads. - [MindsHub in VS Code](https://docs.mindshub.ai/inference/vscode.md): Add MindsHub models to VS Code chat with Copilot's Custom Endpoint provider. - [Web search](https://docs.mindshub.ai/inference/web-search.md): Give the model the web with a built-in tool, on any API, with no search API to sign up for.