V4V MCP 让 AI 助手能够发现生成模型、使用参考媒体、创建图片、视频、音频和聊天响应,并检索结果。它使用与 V4V API 相同的账户、点数和生成历史记录。
服务器 URL:https://v4v.ai/api/mcp
传输方式: 可流式传输的 HTTP
身份验证: 在已启用的部署中使用 OAuth 2.1 登录,或在支持自定义请求头的客户端中使用 API 密钥。两种方式都使用你现有的 V4V 账户和点数。
测试其他环境时,请用你的部署域名替换 v4v.ai。服务器已托管,无需在本地运行进程。
使用 OAuth 登录
Choose this option in ChatGPT, Claude, or a supported editor or CLI. You do not need to create an API key. Add the remote server URL, select OAuth / Sign in, and complete the V4V permission screen. See OAuth authorization and the client-specific steps below.
Add https://v4v.ai/api/mcp as a remote Streamable HTTP server in your assistant.
Choose OAuth / Sign in. If asked for an OAuth client, select Register automatically (dynamic client registration, or DCR). Leave Client ID and Client Secret empty.
Your assistant opens V4V in your browser. Sign in to the V4V account whose credits and library you want to use.
Review the assistant and all three required permissions: view your account and media; upload, estimate costs, edit and share; generate media and transcribe using credits. All three are selected and locked. Choose Accept all permissions, or Cancel to decline.
Return to your assistant, finish connecting if prompted, and ask it to check your balance. Tool permissions and confirmations inside the assistant are managed separately by that client.
{
"model": "chat.gpt_6_astra",
"input": {
"prompt": "Explain this idea in three sentences."
},
"parameters": {
"reasoningEffort": "low",
"webSearch": false,
"stream": false
},
"idempotency_key": "astra-question-001"
}
Optional input.imageUrls accepts up to 12 public URLs or owned file IDs. Text-only requests use string input upstream; images use the model’s multimodal message array. Optional input.functionTools accepts a JSON string of function definitions; web search must be off when functions are present. Function selection is automatic, and completed task results include function_calls for the caller to handle. The model node does not execute functions. Poll get_task using the returned task ID. Completed results entries have type: "text", text, usage, credits and incomplete instead of a download URL.
Omni Flash 1.2 is the default for an unspecified video request; 1.1 remains available by explicit selection. Use get_model for image.text_to_image or ai.image_edit to discover GPT Image 2.6 Flare, Sunburst and Economy variants. Pass parameters.outputFormat and respect allowed_by_variant: Flare/Sunburst text-to-image support PNG/JPEG/WebP/GIF/SVG; their image-to-image variants support PNG/JPEG/WebP; Economy text supports PNG/GIF and Economy image supports PNG. Output downloads preserve GIF frames and SVG vector bytes. SVG conversion usage is included in final billing.
Read get_model.voices for each preset’s exact id, gender, description
and free preview_url. Send the selected ID as parameters.voice; omit
metadata and preview URLs from generation arguments. Check the balance and call
estimate_generation with source="lab" first, then call generate_lab. Read get_task for
owned audio file IDs and signed download links. Delivery instructions belong in
style; they are not spoken. Flash focuses on expressive narration, Flash-Lite
on efficient generation. Provider selection is managed in admin settings, not
through an MCP parameter. ElevenLabs text-to-voice is temporarily disabled.
Kit 0.1.32 adds clipMasks and layeredParallax capability flags. Search kind=effect for mask-circle, mask-ellipse, or mask-rounded-rect, then read get_creative_item for the actual schema and limits. Apply set_mask through edit_editor_project at the saved revision with write access; mask:null clears it. The mask's x/y/width/height are clip-relative fractions and its bounds must remain inside the clip box. It follows rotation/scale/keyframes, clips after effects and before transitions, and does not change audio. Locked layers and audio-only targets reject the operation.