# Token4AI Cloud

> Public customer and developer documentation for model APIs and cloud agents. Use the current key’s model catalog for availability.

## Guides

- [Quickstart: your first model request](https://docs.token4ai.cloud/en/guides/quickstart/): Create an API key, discover an available model, send a request and verify usage in Token4AI Cloud.
  - [中文](https://docs.token4ai.cloud/guides/quickstart/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/quickstart.md)
- [Authentication, API keys and billing groups](https://docs.token4ai.cloud/en/guides/authentication/): Configure Bearer, Anthropic and Gemini credentials; understand key rotation, scopes and billing groups.
  - [中文](https://docs.token4ai.cloud/guides/authentication/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/authentication.md)
- [Models and capability discovery](https://docs.token4ai.cloud/en/guides/models/): List models visible to your key and choose the appropriate model id, protocol and capabilities.
  - [中文](https://docs.token4ai.cloud/guides/models/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/models.md)
- [Chat Completions](https://docs.token4ai.cloud/en/guides/chat-completions/): Use the OpenAI-compatible chat endpoint for messages, conversation context, output controls and streaming.
  - [中文](https://docs.token4ai.cloud/guides/chat-completions/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/chat-completions.md)
- [Responses and conversation state](https://docs.token4ai.cloud/en/guides/responses/): Use the Responses API and understand input, stream, store, previous_response_id, retrieval and deletion.
  - [中文](https://docs.token4ai.cloud/guides/responses/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/responses.md)
- [Anthropic Messages and token estimates](https://docs.token4ai.cloud/en/guides/anthropic/): Configure Anthropic-native requests, system instructions, content blocks, streaming and count_tokens.
  - [中文](https://docs.token4ai.cloud/guides/anthropic/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/anthropic.md)
- [Gemini native content generation](https://docs.token4ai.cloud/en/guides/gemini/): Use generateContent and streamGenerateContent with contents, parts and generationConfig.
  - [中文](https://docs.token4ai.cloud/guides/gemini/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/gemini.md)
- [Streaming and connection handling](https://docs.token4ai.cloud/en/guides/streaming/): Implement SSE parsing, cancellation, timeouts and recovery without duplicate generation or false success states.
  - [中文](https://docs.token4ai.cloud/guides/streaming/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/streaming.md)
- [Tool calling and structured output](https://docs.token4ai.cloud/en/guides/tools/): Build tool-call turns, validate function arguments and use JSON Schema for verifiable model output.
  - [中文](https://docs.token4ai.cloud/guides/tools/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/tools.md)
- [Embeddings and reranking](https://docs.token4ai.cloud/en/guides/embeddings-rerank/): Build retrieval-augmented generation with embeddings and reranking, including input, dimensions and relevance scores.
  - [中文](https://docs.token4ai.cloud/guides/embeddings-rerank/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/embeddings-rerank.md)
- [Vision input and image generation](https://docs.token4ai.cloud/en/guides/multimodal/): Distinguish vision understanding from generation, using image_url content parts and images/generations.
  - [中文](https://docs.token4ai.cloud/guides/multimodal/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/multimodal.md)
- [Audio transcription, translation and speech](https://docs.token4ai.cloud/en/guides/audio/): Use audio/transcriptions, audio/translations and audio/speech with multipart uploads and binary output.
  - [中文](https://docs.token4ai.cloud/guides/audio/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/audio.md)
- [Files API and file lifecycle](https://docs.token4ai.cloud/en/guides/files/): Upload, list, retrieve, download and delete files; distinguish Files API objects, static assets and model input.
  - [中文](https://docs.token4ai.cloud/guides/files/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/files.md)
- [Python, JavaScript and developer tools](https://docs.token4ai.cloud/en/guides/sdk/): Connect official SDKs, custom base URLs, Codex and Claude Code to Token4AI Cloud.
  - [中文](https://docs.token4ai.cloud/guides/sdk/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/sdk.md)
- [Create and run cloud agents](https://docs.token4ai.cloud/en/guides/agents/): Configure Codex, Claude Code, Pi, OpenClaw and Hermes with models, execution environments and tasks.
  - [中文](https://docs.token4ai.cloud/guides/agents/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/agents.md)
- [Room conversations and recurring tasks](https://docs.token4ai.cloud/en/guides/rooms/): Trigger tasks with @Agent, schedule minute or hour loops and stop recurring execution from a Room.
  - [中文](https://docs.token4ai.cloud/guides/rooms/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/rooms.md)
- [Task details, event logs and project boards](https://docs.token4ai.cloud/en/guides/tasks/): Inspect execution state, event duration and failures, and relate agent results to project completion criteria.
  - [中文](https://docs.token4ai.cloud/guides/tasks/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/tasks.md)
- [Private runtimes and cloud execution](https://docs.token4ai.cloud/en/guides/runtimes/): Connect a private device for agent execution and understand availability, project access and harness compatibility.
  - [中文](https://docs.token4ai.cloud/guides/runtimes/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/runtimes.md)
- [Wallet, billing groups and recharge orders](https://docs.token4ai.cloud/en/guides/billing/): Understand USD balances, group pricing, token settlement, recharge credit and order troubleshooting.
  - [中文](https://docs.token4ai.cloud/guides/billing/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/billing.md)
- [Overview, usage history and cost analysis](https://docs.token4ai.cloud/en/guides/usage/): Read daily spend, total tokens, streaming requests and errors, then investigate usage per request.
  - [中文](https://docs.token4ai.cloud/guides/usage/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/usage.md)
- [Static assets and folders](https://docs.token4ai.cloud/en/guides/assets/): Upload, preview, download and organize assets in Vega, including size and access constraints.
  - [中文](https://docs.token4ai.cloud/guides/assets/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/assets.md)
- [Token4AI CLI and agent automation](https://docs.token4ai.cloud/en/guides/cli/): Authorize the CLI in your browser and automate wallet, key, usage and asset operations with JSON output.
  - [中文](https://docs.token4ai.cloud/guides/cli/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/cli.md)
- [Account security and data regions](https://docs.token4ai.cloud/en/guides/security/): Manage sign-in, passkeys, profile and data region settings, including credential and privacy boundaries.
  - [中文](https://docs.token4ai.cloud/guides/security/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/security.md)
- [Support, evidence attachments and incident reports](https://docs.token4ai.cloud/en/guides/support/): Report issues through support chat or tickets, attach images/files and provide actionable request or order identifiers.
  - [中文](https://docs.token4ai.cloud/guides/support/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/support.md)
- [Errors, 429 responses and retry policy](https://docs.token4ai.cloud/en/guides/errors/): Distinguish credential, validation, model, rate-limit and upstream errors, with bounded retries and request ids.
  - [中文](https://docs.token4ai.cloud/guides/errors/)
  - [Markdown](https://docs.token4ai.cloud/markdown/en/errors.md)

## API reference

- [OpenAPI](https://docs.token4ai.cloud/openapi.json): Documented model and file operations; vendor-specific options may vary.