# Crazyrouter API Document

## Docs
- 根目录 > Crazyrouter Overview [Introduction](https://apidocs.crazyrouter.com/en/introduction-7892808m0.md): 
- 根目录 > Crazyrouter Overview [API request](https://apidocs.crazyrouter.com/en/api-request-7892807m0.md): 
- 根目录 > Crazyrouter Overview [API Quick Start Guide](https://apidocs.crazyrouter.com/en/api-quick-start-guide-7892814m0.md): 
- 根目录 > Crazyrouter Overview [API Endpoint](https://apidocs.crazyrouter.com/en/api-endpoint-7892812m0.md): 
- 根目录 > Crazyrouter Overview [Contact us](https://apidocs.crazyrouter.com/en/contact-us-7892809m0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Chat Completion Object](https://apidocs.crazyrouter.com/en/chat-completion-object-7892815m0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Chat Completion Chunk Object](https://apidocs.crazyrouter.com/en/chat-completion-chunk-object-7892816m0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Completions [Completion Object](https://apidocs.crazyrouter.com/en/completion-object-7892817m0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Embeddings [Embedding Object](https://apidocs.crazyrouter.com/en/embedding-object-7892818m0.md): 
- 根目录 > Chat > Anthropic Claude [Chat Completion Object](https://apidocs.crazyrouter.com/en/chat-completion-object-7892819m0.md): 
- 根目录 > Chat > Anthropic Claude [Chat Completion Chunk Object](https://apidocs.crazyrouter.com/en/chat-completion-chunk-object-7892820m0.md): 
- 根目录 > Chat (Responses) [Responses API vs Chat API](https://apidocs.crazyrouter.com/en/responses-api-vs-chat-api-7892821m0.md): 
- 根目录 > Image Generation Models [Image Object](https://apidocs.crazyrouter.com/en/image-object-7892823m0.md): 
- 根目录 > Video Models > Veo > Unified Video Format [Unified Video API Task Status](https://apidocs.crazyrouter.com/en/unified-video-api-task-status-7892827m0.md): 
- 根目录 > Video Models > Kling [Callback Protocol](https://apidocs.crazyrouter.com/en/callback-protocol-7892829m0.md): 
- 根目录 > Video Models > Jimeng Video [Jimeng Task Status](https://apidocs.crazyrouter.com/en/jimeng-task-status-7892831m0.md): 
- 根目录 > Video Models > Hailuo (MiniMax) [Hailuo Task Status](https://apidocs.crazyrouter.com/en/hailuo-task-status-7892833m0.md): 
- 根目录 > Suno Music [Suno Parameters](https://apidocs.crazyrouter.com/en/suno-parameters-7892835m0.md): 
- 根目录 > Suno Music [Suno Overview](https://apidocs.crazyrouter.com/en/suno-overview-7892836m0.md): 
- 根目录 > Python [Python Speech-to-Text (whisper-1)](https://apidocs.crazyrouter.com/en/python-speech-to-text-whisper-1-7892837m0.md): 
- 根目录 > Python [Python Embeddings](https://apidocs.crazyrouter.com/en/python-embeddings-7892839m0.md): 
- 根目录 > Python [Python Function Calling Demo](https://apidocs.crazyrouter.com/en/python-function-calling-demo-7892841m0.md): 
- 根目录 > Python [Python LangChain Demo](https://apidocs.crazyrouter.com/en/python-langchain-demo-7892842m0.md): 
- 根目录 > Python [Python llama_index Configuration](https://apidocs.crazyrouter.com/en/python-llama-index-configuration-7892843m0.md): 
- 根目录 > Python [Python Basic Chat](https://apidocs.crazyrouter.com/en/python-basic-chat-7892844m0.md): 
- 根目录 > Python [Python Local Image Recognition (Multimodal)](https://apidocs.crazyrouter.com/en/python-local-image-recognition-multimodal-7892845m0.md): 
- 根目录 > Python [Python Remote Image Recognition (Multimodal)](https://apidocs.crazyrouter.com/en/python-remote-image-recognition-multimodal-7892846m0.md): 
- 根目录 > Python [Python Image Recognition with Claude](https://apidocs.crazyrouter.com/en/python-image-recognition-with-claude-7892847m0.md): 
- 根目录 > Python [Python Library Streaming Output](https://apidocs.crazyrouter.com/en/python-library-streaming-output-7892848m0.md): 
- 根目录 > Python [Python requests streaming output demo](https://apidocs.crazyrouter.com/en/python-requests-streaming-output-demo-7892850m0.md): 
- 根目录 > Python [Python image generation and editing (gpt-image-2)](https://apidocs.crazyrouter.com/en/python-image-generation-and-editing-gpt-image-2-7892851m0.md): 
- 根目录 > Python [Python official OpenAI library (AutoGPT / LangChain, etc.)](https://apidocs.crazyrouter.com/en/python-official-openai-library-autogpt-langchain-etc-7892852m0.md): 
- 根目录 > Python [Python multi-turn conversation](https://apidocs.crazyrouter.com/en/python-multi-turn-conversation-7892853m0.md): 
- 根目录 > PHP [PHP Image Editing Demo](https://apidocs.crazyrouter.com/en/php-image-editing-demo-7892857m0.md): 
- 根目录 > Node.js [Node.js Audio Transcription Demo](https://apidocs.crazyrouter.com/en/node-js-audio-transcription-demo-7892858m0.md): 
- 根目录 > Plugins & Client Setup Guides [Codex ](https://apidocs.crazyrouter.com/en/codex-7892859m0.md): 
- 根目录 > Plugins & Client Setup Guides [N8N ](https://apidocs.crazyrouter.com/en/n8n-7892860m0.md): 
- 根目录 > Plugins & Client Setup Guides [Gemini CLI ](https://apidocs.crazyrouter.com/en/gemini-cli-7892861m0.md): 
- 根目录 > Plugins & Client Setup Guides [Claude Code Installation and Usage Guide](https://apidocs.crazyrouter.com/en/claude-code-installation-and-usage-guide-7892862m0.md): 
- 根目录 > Plugins & Client Setup Guides [CherryStudio Configuration Guide](https://apidocs.crazyrouter.com/en/cherrystudio-configuration-guide-7892863m0.md): 
- 根目录 > Plugins & Client Setup Guides [dify Configuration Guide](https://apidocs.crazyrouter.com/en/dify-configuration-guide-7892866m0.md): 
- 根目录 > Plugins & Client Setup Guides [cline ](https://apidocs.crazyrouter.com/en/cline-7892867m0.md): 
- 根目录 > Plugins & Client Setup Guides [aider](https://apidocs.crazyrouter.com/en/aider-7892868m0.md): 
- 根目录 > Plugins & Client Setup Guides [Cursor](https://apidocs.crazyrouter.com/en/cursor-7892869m0.md): 
- 根目录 > Plugins & Client Setup Guides [lobechat](https://apidocs.crazyrouter.com/en/lobechat-7892870m0.md): 
- 根目录 > Plugins & Client Setup Guides [ChatBox](https://apidocs.crazyrouter.com/en/chatbox-7892871m0.md): 
- 根目录 > Plugins & Client Setup Guides [gpt_academic](https://apidocs.crazyrouter.com/en/gpt-academic-7892872m0.md): 
- 根目录 > Plugins & Client Setup Guides [nextchat](https://apidocs.crazyrouter.com/en/nextchat-7892873m0.md): 
- 根目录 > Plugins & Client Setup Guides [zotero gpt](https://apidocs.crazyrouter.com/en/zotero-gpt-7892874m0.md): 
- 根目录 > Plugins & Client Setup Guides [CLAUDE DEV ](https://apidocs.crazyrouter.com/en/claude-dev-7892875m0.md): 
- 根目录 > Plugins & Client Setup Guides [ChatGPT Sidebar](https://apidocs.crazyrouter.com/en/chatgpt-sidebar-7892877m0.md): 
- 根目录 > Plugins & Client Setup Guides [chatgpt-on-wechat ](https://apidocs.crazyrouter.com/en/chatgpt-on-wechat-7892878m0.md): 
- 根目录 > Plugins & Client Setup Guides [chatgpt GPT Academic ](https://apidocs.crazyrouter.com/en/chatgpt-gpt-academic-7892879m0.md): 
- 根目录 > Plugins & Client Setup Guides [RikkaHub](https://apidocs.crazyrouter.com/en/rikkahub-7892880m0.md): 
- 根目录 > Plugins & Client Setup Guides [Coze Workflow Configuration Guide](https://apidocs.crazyrouter.com/en/coze-workflow-configuration-guide-7892881m0.md): 
- 根目录 > Help Center [Thinking Model Response Format](https://apidocs.crazyrouter.com/en/thinking-model-response-format-7892882m0.md): 
- 根目录 > Help Center [HTTP Status Codes and Their Meanings](https://apidocs.crazyrouter.com/en/http-status-codes-and-their-meanings-7892883m0.md): 
- 根目录 > Help Center [Image Upload API](https://apidocs.crazyrouter.com/en/image-upload-api-7892884m0.md): 
- [Codex ](https://apidocs.crazyrouter.com/en/codex-7892859m0.md?nav=nav-tutorials): 
- [N8N ](https://apidocs.crazyrouter.com/en/n8n-7892860m0.md?nav=nav-tutorials): 
- [Gemini CLI ](https://apidocs.crazyrouter.com/en/gemini-cli-7892861m0.md?nav=nav-tutorials): 
- [Claude Code Installation and Usage Guide](https://apidocs.crazyrouter.com/en/claude-code-installation-and-usage-guide-7892862m0.md?nav=nav-tutorials): 
- [CherryStudio Configuration Guide](https://apidocs.crazyrouter.com/en/cherrystudio-configuration-guide-7892863m0.md?nav=nav-tutorials): 
- [dify Configuration Guide](https://apidocs.crazyrouter.com/en/dify-configuration-guide-7892866m0.md?nav=nav-tutorials): 
- [cline ](https://apidocs.crazyrouter.com/en/cline-7892867m0.md?nav=nav-tutorials): 
- [aider](https://apidocs.crazyrouter.com/en/aider-7892868m0.md?nav=nav-tutorials): 
- [Cursor](https://apidocs.crazyrouter.com/en/cursor-7892869m0.md?nav=nav-tutorials): 
- [lobechat](https://apidocs.crazyrouter.com/en/lobechat-7892870m0.md?nav=nav-tutorials): 
- [ChatBox](https://apidocs.crazyrouter.com/en/chatbox-7892871m0.md?nav=nav-tutorials): 
- [gpt_academic](https://apidocs.crazyrouter.com/en/gpt-academic-7892872m0.md?nav=nav-tutorials): 
- [nextchat](https://apidocs.crazyrouter.com/en/nextchat-7892873m0.md?nav=nav-tutorials): 
- [zotero gpt](https://apidocs.crazyrouter.com/en/zotero-gpt-7892874m0.md?nav=nav-tutorials): 
- [CLAUDE DEV ](https://apidocs.crazyrouter.com/en/claude-dev-7892875m0.md?nav=nav-tutorials): 
- [ChatGPT Sidebar](https://apidocs.crazyrouter.com/en/chatgpt-sidebar-7892877m0.md?nav=nav-tutorials): 
- [chatgpt-on-wechat ](https://apidocs.crazyrouter.com/en/chatgpt-on-wechat-7892878m0.md?nav=nav-tutorials): 
- [chatgpt GPT Academic ](https://apidocs.crazyrouter.com/en/chatgpt-gpt-academic-7892879m0.md?nav=nav-tutorials): 
- [RikkaHub](https://apidocs.crazyrouter.com/en/rikkahub-7892880m0.md?nav=nav-tutorials): 
- [Coze Workflow Configuration Guide](https://apidocs.crazyrouter.com/en/coze-workflow-configuration-guide-7892881m0.md?nav=nav-tutorials): 

## API Docs
- 根目录 > Chat > ChatGPT > ChatGPT Audio [whisper-1](https://apidocs.crazyrouter.com/en/whisper-1-394334449e0.md): Official docs: https://platform.openai.com/docs/guides/speech-to-text
- 根目录 > Chat > ChatGPT > ChatGPT Audio [Create Translation (whisper-1, audio to English, untested)](https://apidocs.crazyrouter.com/en/create-translation-whisper-1-audio-to-english-untested-394334452e0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Completion (Streaming)](https://apidocs.crazyrouter.com/en/create-chat-completion-streaming-394334453e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Completion (Non-Streaming)](https://apidocs.crazyrouter.com/en/create-chat-completion-non-streaming-394334454e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Vision (Streaming)](https://apidocs.crazyrouter.com/en/create-chat-vision-streaming-394334455e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Vision (Streaming) base64](https://apidocs.crazyrouter.com/en/create-chat-vision-streaming-base64-394334456e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Vision (Non-Streaming)](https://apidocs.crazyrouter.com/en/create-chat-vision-non-streaming-394334457e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Image Generation (Non-Streaming)](https://apidocs.crazyrouter.com/en/create-chat-image-generation-non-streaming-394334458e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Function calling](https://apidocs.crazyrouter.com/en/function-calling-394334459e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Official N Test](https://apidocs.crazyrouter.com/en/official-n-test-394334460e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Function Call](https://apidocs.crazyrouter.com/en/create-chat-function-call-394334462e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Structured Output](https://apidocs.crazyrouter.com/en/create-structured-output-394334463e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Control Reasoning Model Effort Level](https://apidocs.crazyrouter.com/en/control-reasoning-model-effort-level-394334464e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Completion qwen-mt-turbo](https://apidocs.crazyrouter.com/en/create-chat-completion-qwen-mt-turbo-394334465e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Chat [Create Chat Completion DeepSeek V3.1 Thinking Level (Streaming)](https://apidocs.crazyrouter.com/en/create-chat-completion-deepseek-v3-1-thinking-level-streaming-394334466e0.md): 
- 根目录 > Chat > ChatGPT > ChatGPT Chat [DeepSeek OCR Recognition](https://apidocs.crazyrouter.com/en/deepseek-ocr-recognition-394334467e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > ChatGPT > ChatGPT Embeddings [Create Embedding](https://apidocs.crazyrouter.com/en/create-embedding-394334469e0.md): Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms.
- 根目录 > Chat > ChatGPT > Web Search [web search](https://apidocs.crazyrouter.com/en/web-search-394334470e0.md): Reference docs: https://platform.openai.com/docs/api-reference/chat/create
- 根目录 > Chat > ChatGPT > ChatGPT Moderations [Content Moderation (Moderations)](https://apidocs.crazyrouter.com/en/content-moderation-moderations-520505776e0.md): OpenAI Moderations-compatible endpoint that classifies text for safety. The request body is the same as OpenAI's (input is required; model is optional and defaults to text-moderation-latest; omni-moderation-latest is recommended). The upstream response is returned as is.
- 根目录 > Chat > Anthropic Claude [Count Input Tokens count_tokens [Native Format]](https://apidocs.crazyrouter.com/en/count-input-tokens-count-tokens-native-format-520505975e0.md): Estimates the input token count of an Anthropic Messages request before sending it. The request body is identical to POST /v1/messages (model, system, messages, tools, etc.); the response only returns input_tokens.
- 根目录 > Chat > Anthropic Claude [Create Chat Completion (Streaming) [Native Format]](https://apidocs.crazyrouter.com/en/create-chat-completion-streaming-native-format-394334471e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Function Call (Streaming) [Native Format]](https://apidocs.crazyrouter.com/en/create-function-call-streaming-native-format-394334472e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Thinking Chat [Native Format]](https://apidocs.crazyrouter.com/en/create-thinking-chat-native-format-394334473e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Thinking Chat](https://apidocs.crazyrouter.com/en/create-thinking-chat-394334474e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Chat Completion (Streaming)](https://apidocs.crazyrouter.com/en/create-chat-completion-streaming-394334475e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Chat Completion (Non-Streaming)](https://apidocs.crazyrouter.com/en/create-chat-completion-non-streaming-394334476e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Chat Vision (Streaming)](https://apidocs.crazyrouter.com/en/create-chat-vision-streaming-394334477e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [Create Chat Vision (Non-Streaming)](https://apidocs.crazyrouter.com/en/create-chat-vision-non-streaming-394334478e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Anthropic Claude [PDF Support [Native Format]](https://apidocs.crazyrouter.com/en/pdf-support-native-format-394334479e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Gemini Native Format [Create Context Cache cachedContents [Native Format]](https://apidocs.crazyrouter.com/en/create-context-cache-cachedcontents-native-format-520506130e0.md): Gemini Context Caching create endpoint; request/response follow Google's native cachedContents format. Suited to caching long documents or large system prompts, which can then be referenced via the cachedContent field in subsequent generateContent requests to save input tokens.
- 根目录 > Chat > Google Gemini > Gemini Native Format [Text Generation](https://apidocs.crazyrouter.com/en/text-generation-394334481e0.md): Official docs: https://ai.google.dev/gemini-api/docs/text-generation?hl=zh-cn#multi-turn-conversations
- 根目录 > Chat > Google Gemini > Gemini Native Format [Text Generation Streaming](https://apidocs.crazyrouter.com/en/text-generation-streaming-394334482e0.md): Official docs: https://ai.google.dev/gemini-api/docs/text-generation?hl=zh-cn#multi-turn-conversations
- 根目录 > Chat > Google Gemini > Gemini Native Format [Text Generation Thingking+streaming](https://apidocs.crazyrouter.com/en/text-generation-thingkingstreaming-394334483e0.md): Official docs: https://ai.google.dev/gemini-api/docs/text-generation?hl=zh-cn#multi-turn-conversations
- 根目录 > Chat > Google Gemini > Gemini Native Format [Image Generation gemini-2.5-flash-image](https://apidocs.crazyrouter.com/en/image-generation-gemini-2-5-flash-image-394334485e0.md): Official docs: https://ai.google.dev/gemini-api/docs/image-generation?hl=zh-cn#gemini-image-editing
- 根目录 > Chat > Google Gemini > Gemini Native Format [Image Generation gemini-2.5-flash-image Control Aspect Ratio +imageSize](https://apidocs.crazyrouter.com/en/image-generation-gemini-2-5-flash-image-control-aspect-ratio-imagesize-394334487e0.md): Official docs: https://ai.google.dev/gemini-api/docs/image-generation?hl=zh-cn#gemini-image-editing
- 根目录 > Chat > Google Gemini > Gemini Native Format [Image Understanding](https://apidocs.crazyrouter.com/en/image-understanding-394334489e0.md): Official docs: https://ai.google.dev/gemini-api/docs/image-understanding?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Formatted Output](https://apidocs.crazyrouter.com/en/formatted-output-394334490e0.md): Official docs: https://ai.google.dev/gemini-api/docs/structured-output?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Function Call](https://apidocs.crazyrouter.com/en/function-call-394334491e0.md): Official docs: https://ai.google.dev/gemini-api/docs/function-calling?hl=zh-cn&example=meeting
- 根目录 > Chat > Google Gemini > Gemini Native Format [Document Understanding](https://apidocs.crazyrouter.com/en/document-understanding-394334492e0.md): Official docs: https://ai.google.dev/gemini-api/docs/document-processing?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [URL context ](https://apidocs.crazyrouter.com/en/url-context-394334493e0.md): Official docs: https://ai.google.dev/gemini-api/docs/document-processing?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Code Execution](https://apidocs.crazyrouter.com/en/code-execution-394334494e0.md): Official docs: https://ai.google.dev/gemini-api/docs/document-processing?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Video Understanding](https://apidocs.crazyrouter.com/en/video-understanding-394334496e0.md): Official docs: https://ai.google.dev/gemini-api/docs/video-understanding?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [URL context](https://apidocs.crazyrouter.com/en/url-context-394334497e0.md): Official docs: https://ai.google.dev/gemini-api/docs/url-context?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Video Understanding-url Native Format](https://apidocs.crazyrouter.com/en/video-understanding-url-native-format-394334498e0.md): Official docs: https://ai.google.dev/gemini-api/docs/video-understanding?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Audio Understanding](https://apidocs.crazyrouter.com/en/audio-understanding-394334500e0.md): Official docs: https://ai.google.dev/gemini-api/docs/url-context?hl=zh-cn
- 根目录 > Chat > Google Gemini > Gemini Native Format [Text Generation gemini-3-pro-preview:generateContent](https://apidocs.crazyrouter.com/en/text-generation-gemini-3-pro-previewgeneratecontent-394334502e0.md): Official docs: https://ai.google.dev/gemini-api/docs/text-generation?hl=zh-cn#multi-turn-conversations
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Gemini Image Generation API [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/gemini-image-generation-api-chat-compatible-format-394334503e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Chat API [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/chat-api-chat-compatible-format-394334504e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Chat API - Thinking 1 [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/chat-api-thinking-1-chat-compatible-format-394334505e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Chat API - Thinking 2 [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/chat-api-thinking-2-chat-compatible-format-394334506e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Vision API [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/vision-api-chat-compatible-format-394334507e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat > Google Gemini > Chat-compatible Format [Chat + File Reading API [Chat Compatible Format]](https://apidocs.crazyrouter.com/en/chat-file-reading-api-chat-compatible-format-394334508e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Chat (Responses) [Create Function Calling](https://apidocs.crazyrouter.com/en/create-function-calling-394334510e0.md): https://platform.openai.com/docs/api-reference/responses/create
- 根目录 > Chat (Responses) [Create Model Response (Streaming)](https://apidocs.crazyrouter.com/en/create-model-response-streaming-394334511e0.md): https://platform.openai.com/docs/api-reference/responses/create
- 根目录 > Chat (Responses) [Create Model Response (Control Thinking Length)](https://apidocs.crazyrouter.com/en/create-model-response-control-thinking-length-394334512e0.md): https://platform.openai.com/docs/api-reference/responses/create
- 根目录 > Chat (Responses) [web search](https://apidocs.crazyrouter.com/en/web-search-394334513e0.md): https://platform.openai.com/docs/api-reference/responses/create
- 根目录 > Chat (Responses) [Create Model Response (GPT-5 Enable Thinking)](https://apidocs.crazyrouter.com/en/create-model-response-gpt-5-enable-thinking-394334514e0.md): https://platform.openai.com/docs/api-reference/responses/create
- 根目录 > Image Generation Models > Midjourney [Submit Video Task (Image-to-Video)](https://apidocs.crazyrouter.com/en/submit-video-task-image-to-video-520505777e0.md): Generates a video (MJ Video) from a successfully completed Imagine/Upscale task. taskId (the original task ID) is required; the gateway reuses the original task's channel and prompt, and the original task must be in SUCCESS status.
- 根目录 > Image Generation Models > Midjourney [Submit Simple Change Task (String Form)](https://apidocs.crazyrouter.com/en/submit-simple-change-task-string-form-520505781e0.md): Shorthand form of Change: content is "<taskId> <action><index>", e.g. "1743326750223591 U2" upscales image 2, "V3" creates a variation of image 3, and "R" rerolls. Returns content_parse_failed if parsing fails.
- 根目录 > Image Generation Models > Midjourney [Get Task Image (Gateway-Hosted Copy)](https://apidocs.crazyrouter.com/en/get-task-image-gateway-hosted-copy-520505784e0.md): Returns the image binary directly by Midjourney task ID (Content-Type is taken from upstream, default image/jpeg), so it can be used directly in <img src>. **This endpoint does not require an API Key**. If the task does not exist, returns 400 {"error":"midjourney_task_not_found"}.
- 根目录 > Image Generation Models > Midjourney [Submit Edits Task (Inpainting)](https://apidocs.crazyrouter.com/en/submit-edits-task-inpainting-520505976e0.md): Midjourney edit (EDITS) action: inpaint / outpaint an already generated image. The request body follows the midjourney-proxy modal-style fields (taskId, prompt, maskBase64, etc., validated upstream).
- 根目录 > Image Generation Models > Midjourney [Submit Shorten Task (Prompt Shortening)](https://apidocs.crazyrouter.com/en/submit-shorten-task-prompt-shortening-520505981e0.md): Midjourney /shorten: analyzes and shortens a prompt, returning suggested versions. prompt is required. Only supported by midjourney-proxy-plus upstreams.
- 根目录 > Image Generation Models > Midjourney [Submit Change Task (U/V/R Actions)](https://apidocs.crazyrouter.com/en/submit-change-task-uvr-actions-520506131e0.md): Performs upscale (UPSCALE), variation (VARIATION), or reroll (REROLL) on a four-grid result. Required: taskId, action, index (1–4; optional for REROLL but must be non-zero).
- 根目录 > Image Generation Models > Midjourney [InsightFace Face Swap](https://apidocs.crazyrouter.com/en/insightface-face-swap-520506133e0.md): Swaps the face in sourceBase64 onto the targetBase64 image (InsightFace). Both fields are required and accept a data URL or raw base64. Billed at the fixed price of model swap_face.
- 根目录 > Image Generation Models > Midjourney [upload image](https://apidocs.crazyrouter.com/en/upload-image-394334516e0.md): Official docs: https://docs.midjourney.com/hc/en-us/articles/33329380893325-Managing-Image-Uploads
- 根目录 > Image Generation Models > Midjourney [upload Imagine task](https://apidocs.crazyrouter.com/en/upload-imagine-task-394334517e0.md): Official docs: https://docs.midjourney.com/hc/en-us/articles/32023408776205-Prompt-Basics
- 根目录 > Image Generation Models > Midjourney [Query Task Status by Task ID](https://apidocs.crazyrouter.com/en/query-task-status-by-task-id-394334518e0.md): 
- 根目录 > Image Generation Models > Midjourney [Query Tasks by ID List](https://apidocs.crazyrouter.com/en/query-tasks-by-id-list-394334519e0.md): 
- 根目录 > Image Generation Models > Midjourney [Get Task Image Seed](https://apidocs.crazyrouter.com/en/get-task-image-seed-394334520e0.md): 
- 根目录 > Image Generation Models > Midjourney [Execute Action](https://apidocs.crazyrouter.com/en/execute-action-394334521e0.md): Official docs: https://docs.midjourney.com/hc/en-us/articles/32804058614669-Upscalers
- 根目录 > Image Generation Models > Midjourney [Submit Blend Task](https://apidocs.crazyrouter.com/en/submit-blend-task-394334522e0.md): Official docs: https://docs.midjourney.com/hc/en-us/articles/32635189884557-Blend-Images-on-Discord
- 根目录 > Image Generation Models > Midjourney [Sumit Describe task](https://apidocs.crazyrouter.com/en/sumit-describe-task-394334523e0.md): Official docs: https://docs.midjourney.com/hc/en-us/articles/32497889043981-Describe
- 根目录 > Image Generation Models > Midjourney [Sumit Modal](https://apidocs.crazyrouter.com/en/sumit-modal-394334524e0.md): 
- 根目录 > Image Generation Models > GPT Image (gpt-image-2) [Create gpt-image-1](https://apidocs.crazyrouter.com/en/create-gpt-image-1-394334534e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Image Generation Models > GPT Image (gpt-image-2) [Create Image (gpt-image-2, OpenAI images/generations format)](https://apidocs.crazyrouter.com/en/create-image-gpt-image-2-openai-imagesgenerations-format-394334540e0.md): [Images](https://platform.openai.com/docs/api-reference/images)
- 根目录 > Image Generation Models > GPT Image (gpt-image-2) [Edit Image (gpt-image-2, mask / multi-image reference)](https://apidocs.crazyrouter.com/en/edit-image-gpt-image-2-mask-multi-image-reference-394334536e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Image Generation Models > Jimeng Image [Create Image](https://apidocs.crazyrouter.com/en/create-image-394334537e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Image Generation Models > Jimeng Image [Edit Image](https://apidocs.crazyrouter.com/en/edit-image-394334538e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-3-0-t2i-250415](https://apidocs.crazyrouter.com/en/doubao-seedream-3-0-t2i-250415-394334544e0.md): Given a prompt and/or an input image, the model generates a new image.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seededit-3-0-i2i-250628](https://apidocs.crazyrouter.com/en/doubao-seededit-3-0-i2i-250628-394334545e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-0-250828 Text-to-Image](https://apidocs.crazyrouter.com/en/doubao-seedream-4-0-250828-text-to-image-394334546e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-0-250828 Image-to-Image](https://apidocs.crazyrouter.com/en/doubao-seedream-4-0-250828-image-to-image-394334547e0.md): Given a prompt and/or an input image, the model generates a new image.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-0-250828 Multi-Image to Image](https://apidocs.crazyrouter.com/en/doubao-seedream-4-0-250828-multi-image-to-image-394334548e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Text-to-Image (text-only input, single image output)](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-text-to-image-text-only-input-single-image-output-394334549e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Text+Image to Image (single image in, single image out)](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-textimage-to-image-single-image-in-single-image-out-394334550e0.md): Given a prompt and/or an input image, the model generates a new image.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Multi-Image Fusion (multiple inputs, single output)](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-multi-image-fusion-multiple-inputs-single-output-394334551e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Image Set Output (multi-image output)](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-image-set-output-multi-image-output-394334552e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Single Image to Image Set](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-single-image-to-image-set-394334553e0.md): Given a prompt and/or an input image, the model generates a new image.
- 根目录 > Image Generation Models > Doubao Seedream [doubao-seedream-4-5-251128 Multi-Reference to Image Set](https://apidocs.crazyrouter.com/en/doubao-seedream-4-5-251128-multi-reference-to-image-set-394334554e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Image Generation Models > Qwen-Image [qwen-image-edit-2509](https://apidocs.crazyrouter.com/en/qwen-image-edit-2509-394334558e0.md): Given a prompt and/or input images, the model generates new images.
- 根目录 > Video Models > Veo > Unified Video Format [Create Video with Image](https://apidocs.crazyrouter.com/en/create-video-with-image-394334560e0.md): 
- 根目录 > Video Models > Veo > Unified Video Format [Create Video (Reference Image)](https://apidocs.crazyrouter.com/en/create-video-reference-image-394334562e0.md): 
- 根目录 > Video Models > Kling [Omni Video omni-video (kling-v3-omni / kling-video-o1)](https://apidocs.crazyrouter.com/en/omni-video-omni-video-kling-v3-omni-kling-video-o1-520505982e0.md): Mirror of the official Kling omni-video endpoint, **only for the kling-v3-omni and kling-video-o1 models** (these two models are rejected on the regular text2video / image2video paths, and conversely other models return 400 on this path). The request body matches the official Kling API; the gateway passes the entire raw body through in metadata.
- 根目录 > Video Models > Kling [Image Generation](https://apidocs.crazyrouter.com/en/image-generation-394334570e0.md): For other parameters, see the official docs: https://app.klingai.com/cn/dev/document-api/apiReference/updateNotice
- 根目录 > Video Models > Kling [Text-to-Video](https://apidocs.crazyrouter.com/en/text-to-video-394334571e0.md): For other parameters, see the official docs: https://app.klingai.com/cn/dev/document-api/apiReference/updateNotice
- 根目录 > Video Models > Kling [Image to Video](https://apidocs.crazyrouter.com/en/image-to-video-394334572e0.md): For other parameters see the official docs: https://app.klingai.com/cn/dev/document-api/apiReference/updateNotice
- 根目录 > Video Models > Kling [Query Task (free)](https://apidocs.crazyrouter.com/en/query-task-free-394334573e0.md): 
- 根目录 > Video Models > Kling [Multi-Image Reference to Video](https://apidocs.crazyrouter.com/en/multi-image-reference-to-video-394334574e0.md): For other parameters, see the official docs: https://app.klingai.com/cn/dev/document-api/apiReference/updateNotice
- 根目录 > Video Models > Kling [Lip Sync](https://apidocs.crazyrouter.com/en/lip-sync-394334576e0.md): 1. Generate a video.
- 根目录 > Video Models > Kling [Extend Video](https://apidocs.crazyrouter.com/en/extend-video-394334577e0.md): 1. Generate a video
- 根目录 > Video Models > Kling [Video Effects](https://apidocs.crazyrouter.com/en/video-effects-394334578e0.md): Apply different effects via effect_scene. Currently supported:
- 根目录 > Video Models > Jimeng Video [Submit Video Generation Task](https://apidocs.crazyrouter.com/en/submit-video-generation-task-394334581e0.md): 
- 根目录 > Video Models > Jimeng Video [Query Video Task (free)](https://apidocs.crazyrouter.com/en/query-video-task-free-394334582e0.md): TaskStatus:
- 根目录 > Video Models > Hailuo (MiniMax) [Submit Video Generation Task](https://apidocs.crazyrouter.com/en/submit-video-generation-task-394334585e0.md): Official docs: https://www.minimax.io/platform/document/Model%3Fkey=684261f14c5738213294faa7?key=66d1439376e52fcee2853049&document=video_generation
- 根目录 > Video Models > Hailuo (MiniMax) [Query Video Task Status](https://apidocs.crazyrouter.com/en/query-video-task-status-394334586e0.md): 
- 根目录 > Video Models > Hailuo (MiniMax) [Image-to-Video](https://apidocs.crazyrouter.com/en/image-to-video-394334587e0.md): Official docs: https://www.minimax.io/platform/document/Model%3Fkey=684261f14c5738213294faa7?key=66d1439376e52fcee2853049&document=video_generation
- 根目录 > Video Models > Doubao Seedance [Text-to-Video Example](https://apidocs.crazyrouter.com/en/text-to-video-example-394334589e0.md): Official docs: https://www.volcengine.com/docs/82379/1520757
- 根目录 > Video Models > Doubao Seedance [Image to Video - First Frame](https://apidocs.crazyrouter.com/en/image-to-video-first-frame-394334590e0.md): Official docs: https://www.volcengine.com/docs/82379/1520757
- 根目录 > Video Models > Doubao Seedance [seedance-lite First/Last Frame](https://apidocs.crazyrouter.com/en/seedance-lite-firstlast-frame-394334591e0.md): 
- 根目录 > Video Models > Doubao Seedance [Image-to-Video (base64 encoded)](https://apidocs.crazyrouter.com/en/image-to-video-base64-encoded-394334592e0.md): Official docs: https://www.volcengine.com/docs/82379/1520757
- 根目录 > Video Models > Doubao Seedance [Query Single Task](https://apidocs.crazyrouter.com/en/query-single-task-394334594e0.md): 
- 根目录 > Video Models > Sora > OpenAI Video Format [OpenAI Query Task](https://apidocs.crazyrouter.com/en/openai-query-task-394334606e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Video Models > Sora > OpenAI Video Format [OpenAI Download Video](https://apidocs.crazyrouter.com/en/openai-download-video-394334607e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Video Models > Sora > OpenAI Video Format [OpenAI Create Video (with image)](https://apidocs.crazyrouter.com/en/openai-create-video-with-image-394334608e0.md): 
- 根目录 > Video Models > Sora > OpenAI Video Format [Create Video from Storyboard](https://apidocs.crazyrouter.com/en/create-video-from-storyboard-394334609e0.md): 
- 根目录 > Video Models > Sora > OpenAI Video Format [OpenAI Create Video with Image (Private Mode)](https://apidocs.crazyrouter.com/en/openai-create-video-with-image-private-mode-394334610e0.md): 
- 根目录 > Video Models > Sora > OpenAI Video Format [OpenAI Create Video (with Character)](https://apidocs.crazyrouter.com/en/openai-create-video-with-character-394334611e0.md): 
- 根目录 > Video Models > Tongyi Wanxiang [Generate Video](https://apidocs.crazyrouter.com/en/generate-video-394334617e0.md): 
- 根目录 > Video Models > Tongyi Wanxiang [Query Video](https://apidocs.crazyrouter.com/en/query-video-394334618e0.md): 
- 根目录 > Video Models > Unified Video API (all video models) [Query Video Task (Unified Format)](https://apidocs.crazyrouter.com/en/query-video-task-unified-format-520505787e0.md): Queries a task submitted via the unified video endpoint by task ID. Equivalent forms: GET /v1/video/create/{task_id}, GET /v1/video/query?id={task_id}.
- 根目录 > Video Models > Unified Video API (all video models) [Submit Video Generation Task (Unified Format)](https://apidocs.crazyrouter.com/en/submit-video-generation-task-unified-format-520506135e0.md): Crazyrouter unified video endpoint: a single request body covers every connected video model (veo, Sora, Kling, Jimeng, Doubao Seedance, Hailuo, Tongyi Wanxiang, Grok, etc.); the gateway routes to the corresponding vendor by model and converts the parameters automatically.
- 根目录 > System API [List Tokens](https://apidocs.crazyrouter.com/en/list-tokens-394334620e0.md): 
- 根目录 > System API [Create Token](https://apidocs.crazyrouter.com/en/create-token-394334621e0.md): 
- 根目录 > System API [Get Models Supported by Token](https://apidocs.crazyrouter.com/en/get-models-supported-by-token-394334622e0.md): 
- 根目录 > System API [Get Account Info](https://apidocs.crazyrouter.com/en/get-account-info-394334623e0.md): 
- 根目录 > System API [Update Token](https://apidocs.crazyrouter.com/en/update-token-394334624e0.md): 
- 根目录 > System API [Delete Token](https://apidocs.crazyrouter.com/en/delete-token-394334626e0.md): 
- 根目录 > System API [Search Tokens](https://apidocs.crazyrouter.com/en/search-tokens-520505839e0.md): 
- 根目录 > System API [Query Async Task (Image / Video)](https://apidocs.crazyrouter.com/en/query-async-task-image-video-520505996e0.md): Generic async task query: applies to image tasks submitted asynchronously (when /v1/images/generations returns a task object) and to video tasks. GET /v1/images/tasks/{task_id} is a legacy alias of this endpoint with identical behavior.
- 根目录 > System API [Get Single Model Info](https://apidocs.crazyrouter.com/en/get-single-model-info-520506139e0.md): Returns information for a single model available to the current token, looked up by model name (OpenAI /v1/models/{model} format). When both the x-api-key and anthropic-version request headers are sent, the Anthropic format is returned (id, created_at, display_name, type).
- 根目录 > Suno Music > Submit Tasks [Generate Song (Inspiration Mode)](https://apidocs.crazyrouter.com/en/generate-song-inspiration-mode-394334627e0.md): 
- 根目录 > Suno Music > Submit Tasks [Generate Song (Custom Mode)](https://apidocs.crazyrouter.com/en/generate-song-custom-mode-394334628e0.md): 
- 根目录 > Suno Music > Submit Tasks [Generate Song (Extend Mode)](https://apidocs.crazyrouter.com/en/generate-song-extend-mode-394334629e0.md): 
- 根目录 > Suno Music > Submit Tasks [Generate Song (Artist Style)](https://apidocs.crazyrouter.com/en/generate-song-artist-style-394334630e0.md): # Integration Steps
- 根目录 > Suno Music > Submit Tasks [Generate Song (Upload Song for Re-creation)](https://apidocs.crazyrouter.com/en/generate-song-upload-song-for-re-creation-394334631e0.md): 
- 根目录 > Suno Music > Submit Tasks [Generate Lyrics](https://apidocs.crazyrouter.com/en/generate-lyrics-394334633e0.md): 
- 根目录 > Suno Music > Submit Tasks [Concatenate Songs](https://apidocs.crazyrouter.com/en/concatenate-songs-394334634e0.md): 
- 根目录 > Suno Music > Query Tasks [Batch Get Tasks](https://apidocs.crazyrouter.com/en/batch-get-tasks-394334635e0.md): 
- 根目录 > Suno Music > Query Tasks [Query Single Task](https://apidocs.crazyrouter.com/en/query-single-task-394334636e0.md): 
- 根目录 > Suno Music > Query Tasks [Get WAV](https://apidocs.crazyrouter.com/en/get-wav-394334637e0.md): 
- 根目录 > Rerank [Rerank](https://apidocs.crazyrouter.com/en/rerank-394334638e0.md): Given a prompt, the model returns one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
- 根目录 > Decisions API (TypeSafe JEV) [Create Decisions](https://apidocs.crazyrouter.com/en/create-decisions-520506005e0.md): OpenRouter Decisions API compatible endpoint for the typesafe/jev-* family of decision models (e.g. jev-1.13): given state (the state/text to be judged) and questions (a number of noul / choice / score type questions), the model returns structured answers. The request body is passed through byte-for-byte in the native OpenRouter format; only model is rewritten according to the channel mapping.
- 根目录 > 3D Generation (Hunyuan hy-3d) [Query 3D Generation Task](https://apidocs.crazyrouter.com/en/query-3d-generation-task-520505848e0.md): Queries the status of a 3D task. status is normalized to queued / processing / succeeded / failed; on success, data.url is the model file URL and data.format is glb / obj / zip. Queries are free.
- 根目录 > 3D Generation (Hunyuan hy-3d) [Submit 3D Generation Task](https://apidocs.crazyrouter.com/en/submit-3d-generation-task-520506160e0.md): Text-to-3D / image-to-3D model generation (Tencent Hunyuan hy-3d series: hy-3d-3.1, hy-3d-3.0, hy-3d-component). The request body follows the unified task format; any of image / images[0] / input_reference is passed upstream as the reference image image_url; fields inside input and metadata are passed through as-is.
- 根目录 > File Upload > Temporary Images [Temporary Base64 Image Upload](https://apidocs.crazyrouter.com/en/temporary-base64-image-upload-520505842e0.md): Uploads a local image and returns a temporary public URL for use with image-to-image, image-to-video, and vision understanding endpoints.
- 根目录 > File Upload > Temporary Images [Get Presigned Direct Upload URL (recommended for large files)](https://apidocs.crazyrouter.com/en/get-presigned-direct-upload-url-recommended-for-large-files-520505999e0.md): First request a presigned URL from this endpoint, then have the client **PUT the file directly to object storage** without going through Crazyrouter servers. Suitable for larger images or poor network conditions.
- 根目录 > File Upload > Temporary Images [Re-host Remote Image as Temporary URL](https://apidocs.crazyrouter.com/en/re-host-remote-image-as-temporary-url-520506001e0.md): Upload a local image and obtain a temporary public URL for use with the image-to-image, image-to-video, and vision understanding endpoints.
- 根目录 > File Upload > Temporary Images [Temporary Local Image Upload](https://apidocs.crazyrouter.com/en/temporary-local-image-upload-520506155e0.md): Upload a local image and get a temporary public URL for use with image-to-image, image-to-video, and vision understanding endpoints.
- 根目录 > File Upload > Playground [Playground Image Upload](https://apidocs.crazyrouter.com/en/playground-image-upload-520506158e0.md): Image upload endpoint for this site's Playground. Source of truth in code: router/api-router.go and controller/playground_upload.go. Requires a logged-in console user session; does not use customer API keys. 20MB per file; returns the public URL, key, file_name, mime_type, and size.
- 根目录 > File Upload > Deprecated [Legacy Upload Endpoint (Unsupported)](https://apidocs.crazyrouter.com/en/legacy-upload-endpoint-unsupported-520505845e0.md): POST /api/upload is not registered in the local code. Do not use this legacy path. For temporary image uploads with a customer API Key, use POST /v1/files/uploads, POST /v1/files/uploads/base64, or POST /v1/files/uploads/url. This site's PicGo image-hosting workflow uses POST /api/picgo/upload; the console Playground uses POST /api/playground/uploads/image.