# UseAPI.net — third-party API for popular AI services

> UseAPI.net provides API access to leading AI models for video creation, image generation, music composition, speech synthesis, face swapping, and more — all through a single $15/month subscription that covers every API. Each API connects your own AI website account(s) to a fully-featured REST API at website subscription prices, significantly cheaper than official API rates. Multi-account configuration with automated load balancing is included. Many of them also work on the service's free plan: Google Flow (50 Flow credits a day for Veo 3.1 Fast and Lite, images at no credit cost), Gemini Notebook (NotebookLM), Muse, Dreamina (Seedream 4.6 images), PixVerse (PixVerse models up to 540p), Mureka (4 songs a day), Flow Music (daily top-up credits), MiniMax Audio (songs and speech), InsightFaceSwap (10 credits a day), Kling (text-to-speech and face detection) and Runway (image upscaler and image descriptions). A free Google account makes only avatars in Google Vids, and TemPolor has no free plan.

<small>Index generated: 2026-10-25 20:06 UTC</small>

## Start here

- [Model Matrix](https://useapi.net/model-matrix): **Canonical list of every model across every API** (video, image, music, speech, face swap, upscale, extend, effects). The authoritative source for what models are available where, with notes on unlimited / free / Premium-only tiers. Always check this first to find which API exposes a given model.
- [Complete API spec](https://useapi.net/assets/aibot/all.txt): Single combined documentation file for all supported APIs (~4 MB). Feed this to your LLM for cross-service integrations. For smaller-context models, fetch only the per-service files listed below.
- [Account management](https://useapi.net/assets/aibot/account-management.txt): Programmatic account endpoints — check credits & usage statistics (`GET account`, `GET account/stats`), set account-level defaults like the global `replyUrl` webhook (`POST account`), reset the API token after an email confirmation (`POST account/token/reset`), and read the daily anonymized stats reports — Google Flow captcha, Google Flow, PixVerse, Muse (`GET stats/daily`, no token).
- [Subscription & pricing](https://useapi.net/docs/subscription): The single flat $15/month useapi.net subscription — what's included and how it works (you connect your own account with each AI service, and many work on its free plan).
- [Integrations](https://useapi.net/docs/postman-swagger): Postman collections for every API, n8n and other automation tools, MCP clients, and OpenAPI notes.
- [Support](https://useapi.net/docs/support): How to get help — Discord, Telegram, and email.

## MCP servers

Hosted Model Context Protocol servers for Claude Code, Cursor, VS Code and other MCP clients. Each uses the useapi.net API token as `Authorization: Bearer <token>` over Streamable HTTP, needs nothing installed, and is listed in the official MCP Registry under `net.useapi`. Overview: [Model Matrix & MCP](https://useapi.net/model-matrix#mcp-servers).

- [Google Flow MCP server](https://useapi.net/docs/api-google-flow-v1/mcp): `https://api.useapi.net/v1/google-flow/mcp`, 12 tools, chat with Google Flow Agent
- [Google Vids MCP server](https://useapi.net/docs/api-google-vids-v1/mcp): `https://api.useapi.net/v1/google-vids/mcp`, 16 tools, Gemini Omni video and voiced avatars
- [Muse MCP server](https://useapi.net/docs/api-muse-v1/mcp): `https://api.useapi.net/v1/muse/mcp`, 37 tools, Meta Muse images, video and chat
- [Gemini Notebook MCP server (NotebookLM MCP server)](https://useapi.net/docs/api-gemini-notebook-v1/mcp): `https://api.useapi.net/v1/gemini-notebook/mcp`, 57 tools
- [Kling MCP server](https://useapi.net/docs/api-kling-v1/mcp): `https://api.useapi.net/v1/kling/mcp`, 45 tools
- [Mureka MCP server](https://useapi.net/docs/api-mureka-v1/mcp): `https://api.useapi.net/v1/mureka/mcp`, 40 tools, songs, stems, sheet music, speech
- [Flow Music MCP server](https://useapi.net/docs/api-flowmusic-v1/mcp): `https://api.useapi.net/v1/flowmusic/mcp`, 18 tools, Lyria songs

## Universal conventions

**Authentication.** Every useapi.net request uses header `Authorization: Bearer <token>`. The token format is `user:<number>-<unique-string>` (e.g. `user:12345-abcdefghi`). **Use the complete string** — include the `user:` prefix and the alphanumeric suffix. Do not truncate to just the number. Do not URL-encode. A single token authorizes every API under the user's subscription.

**Cross-service shared models.** The same underlying model (Veo 3.1, Seedance 2.0, Nano Banana, etc.) is often exposed by multiple useapi.net APIs. Request and response shapes DIFFER between APIs even for the same model — each API exposes its own native interface. Veo 3.1 via Google Flow uses a different request body than Veo 3.1 via Runway. See the Model Matrix to find which API exposes a model, then consult THAT API's documentation for its specific format.

**Everything else is service-specific.** Identifier names (`jobid` / `taskId` / `musicId` / etc.), job lifecycle, response shapes, webhook payload shape, status code semantics, and synchronous-vs-asynchronous behavior vary per API. Use the per-service documentation below as the authoritative source for each API.

## Tutorials & Guides

Worked, copy-pasteable tutorials (curl commands, runnable scripts, and sample output) for common tasks. Combined full text of every guide: [all tutorials (articles.txt)](https://useapi.net/assets/aibot/articles.txt). Individual guides by API:

**Cross-API comparisons**
- [Suno API Alternative: Mureka, Lyria 3.5, TemPolor and MiniMax](https://useapi.net/docs/articles/suno-api-alternative): full songs with vocals over REST from Mureka V9.5, Lyria 3.5, TemPolor v4.8 and MiniMax Music 3.0 on a Mureka Premier, Flow Music Member, TemPolor Ultra or MiniMax Audio Pro plan, the price per song against each official API, and real tracks
- [Sora 2 API Alternatives Compared, With Code: Seedance 2.5, Omni Flash, MiniMax H3, Kling v3](https://useapi.net/docs/articles/sora-2-api-alternatives)
- [Four AI Music APIs on the Same Six Briefs: TemPolor v4.8, MiniMax Music-3.0, Lyria 3.5 and Mureka V9.5](https://useapi.net/docs/articles/ai-music-flagships-compared)
- [Seedance 2.0 and 2.5 API Pricing: Every Route, Compared](https://useapi.net/docs/articles/seedance-2-api-pricing)
- [AI Music APIs Compared: Mureka vs MiniMax vs ElevenLabs vs Lyria 3 Pro](https://useapi.net/docs/articles/ai-music-api-comparison)

**Google Flow**
- [How to Chat with Google Flow Agent via the Google Flow API](https://useapi.net/docs/articles/google-flow-agent-chat)
- [Google Flow Errors Explained: Upscaling Failed, PUBLIC_ERROR_UNUSUAL_ACTIVITY and Unsafe Generation](https://useapi.net/docs/articles/google-flow-errors)
- [Nano Banana 2 Lite vs 2.1 vs Pro: Google Flow Image Models Compared](https://useapi.net/docs/articles/google-flow-nano-banana-compare)
- [Veo 3.1 API Pricing: Official Google Veo API vs Google Flow API](https://useapi.net/docs/articles/veo-3-1-api-pricing): Google's per-second Veo 3.1 rates (Gemini API Preview shuts down 2026-10-22, Agent Platform same rates) against your own Google AI Pro / Ultra plan through useapi.net, price per video on Pro and Ultra subscriptions and a real run
- [How to Use the Veo 3.1 API: Generate Video with curl via the Google Flow API](https://useapi.net/docs/articles/google-flow-bash)
- [How to Generate Nano Banana 2.1 & Nano Banana Pro Images via the Google Flow API](https://useapi.net/docs/articles/google-flow-images-bash)
- [如何通过 Google Flow API 调用 Veo 3.1、Omni 1.1 Flash 和 Nano Banana Pro（中文教程）](https://useapi.net/docs/articles/google-flow-api-zh) (in Simplified Chinese)
- [How to Generate Audio-Native AI Video with Gemini Omni Flash via the Google Flow API](https://useapi.net/docs/articles/omni-flash-bash)
- [How to Make a UGC Product Video with the Google Flow API](https://useapi.net/docs/articles/google-flow-ugc-product-video)
- [How to Make UGC Video Ads in n8n with the Google Flow API](https://useapi.net/docs/articles/google-flow-n8n-ugc-ads)

**Google Vids**
- [How to Make AI Influencer Videos with Avatars via the Google Vids API](https://useapi.net/docs/articles/google-vids-influencer-avatars)
- [如何通过 Google Vids API 用 Gemini Omni 1.1 Flash 生成会说话的数字人视频（中文教程）](https://useapi.net/docs/articles/google-vids-api-zh) (in Simplified Chinese): voiced avatars, extends up to about 41 s, no captcha, Vids allowance vs Flow credits, curl steps

**Muse**
- [Is There a Meta Muse Video API? Official API, Free Tools and a Hosted API (2026)](https://useapi.net/docs/articles/meta-muse-video-api): Meta's Model API (Muse Image, Muse Wonder, Muse Spark, no video model) vs the self-hosted muse2api vs per-request resellers vs the hosted useapi.net Muse API, a comparison table, one curl for a 10-second video, the MCP one-liner, FAQ
- [How to Generate AI Videos and Images with the Meta Muse API](https://useapi.net/docs/articles/muse-images-videos): text to image, combining several photos, old-photo restoration, image to video with dialogue, swapping the person in a video, chaining generated files, the free weekly allowance
- [如何通过 Muse API 用 Meta Muse 生成 AI 图片和带对白的视频（中文教程）](https://useapi.net/docs/articles/muse-api-zh) (in Simplified Chinese): the same walkthrough as the English tutorial, multi-photo composite, old-photo restoration, image to video with dialogue, video person swap, free weekly allowance, curl steps

**Gemini Notebook (NotebookLM)**
- [Gemini Notebook (NotebookLM) API Options in 2026: Google Enterprise API, Free Libraries or a Hosted API](https://useapi.net/docs/articles/notebooklm-api-options): what each option offers, with a comparison table, and how to start with the hosted REST API or MCP server.
- [How to Automate Gemini Notebook (NotebookLM) with the API: Deep Research, Chat, Audio Overviews and a Lot More](https://useapi.net/docs/articles/gemini-notebook-bash)
- [Turn a Web Page or YouTube Video into a Podcast with Gemini Notebook (NotebookLM): A Real Example](https://useapi.net/docs/articles/notebooklm-podcast-example)
- [How to Turn Web Pages, YouTube Videos and RSS Feeds into Gemini Notebook (NotebookLM) Podcasts with n8n](https://useapi.net/docs/articles/notebooklm-n8n-podcast)
- [如何用 Gemini Notebook API（NotebookLM API）把网页、PDF 和 YouTube 视频变成中文播客（中文教程）](https://useapi.net/docs/articles/gemini-notebook-api-zh) (in Simplified Chinese): one-call Audio Overview with language zh_Hans, notebook flow, Python, scaling across accounts.

**Flow Music**
- [How to Generate AI Music with Lyria 3 Pro via the Flow Music API](https://useapi.net/docs/articles/flowmusic-bash)

**Dreamina**
- [How to Generate AI Video with Seedance 2.5 and 2.0 via the Dreamina API](https://useapi.net/docs/articles/dreamina-bash)

**Kling**
- [How to Generate Kling 4.0 Video via the Kling API](https://useapi.net/docs/articles/kling-4-api)
- [How to Generate AI Video with Kling v3 via the Kling API](https://useapi.net/docs/articles/kling-bash)
- [How to Generate Multi-Reference & Multi-Shot Video with Kling Omni via the Kling API](https://useapi.net/docs/articles/kling-omni-bash)
- [Kling Lip-Sync Avatars, Motion Control & Image Generation via the Kling API](https://useapi.net/docs/articles/kling-features-bash)

**Runway**
- [How to Generate Runway Gen Videos via API](https://useapi.net/docs/articles/runway-bash)
- [How to Batch-Generate Images with Runway Frames via API](https://useapi.net/docs/articles/runway-frames-script)

**PixVerse**
- [How to Generate AI Video with PixVerse V6 via the PixVerse API](https://useapi.net/docs/articles/pixverse-demo)

**MiniMax / Hailuo**
- [How to Generate Hailuo 02 and 2.3 Video with curl: MiniMax Tutorial](https://useapi.net/docs/articles/minimax-bash)
- [How to Generate Text-to-Speech (TTS) Audio with the MiniMax (Hailuo) API](https://useapi.net/docs/articles/minimax-speech-bash): Speech 2.8 end to end — voices, emotion/sound/pause markup, speed and pitch, word timings for captions, voice cloning and SSE streaming, with playable samples.

**Mureka**
- [How to Turn a Song or YouTube Link into Sheet Music with the Mureka API](https://useapi.net/docs/articles/mureka-sheet-music): POST /music/score end to end from a song, a YouTube link or an uploaded MP3, the zip (MusicXML, staff PDF, numbered-notation PDF), opening it in MuseScore, limits and the 10 Mureka Gold cost, plus the MCP route.
- [Mureka API Music Generation](https://useapi.net/docs/articles/mureka-music-samples)
- [How to Generate AI Music with the Mureka API](https://useapi.net/docs/articles/mureka-bash)

**Face Swap**
- [How to Face Swap & Animate AI Images via API](https://useapi.net/docs/articles/imagine-faceswap-animate)

## Blog & Announcements

Chronological feature, model, and API announcements (a "what's new" history across all services). Combined full text: [all blog posts (blog.txt)](https://useapi.net/assets/aibot/blog.txt). Browse: [Blog](https://useapi.net/blog).

## Google Flow API

Runs on your [Google AI subscription](https://gemini.google/subscriptions/): Free, Plus $4.99, Pro $19.99, Ultra $99.99 or Ultra $199.99 a month (as of October 2026).

- [Google Flow API documentation](https://useapi.net/assets/aibot/api-google-flow-v1.txt): Third-party API for Google's [Flow](https://flow.google.com) (Veo + Gemini Omni Flash + Nano Banana). **Video**: Veo 3.1 Quality, Veo 3.1 Fast, Veo 3.1 Lite, Veo 3.1 Lite Lower Priority (0 credits, only for the owner of an Ultra $199.99 plan, not the family members the owner invites), Omni 1.1 Flash / Gemini Omni Flash (Plus, Pro or Ultra — audio-native T2V/R2V at 4/6/8/10s in 360p or 720p, start+end frames, V2V edit on uploaded MP4 via `referenceVideo_1`, up to 7 image refs + 5 voice presets for spoken dialogue). **Images**: Nano Banana 2 Lite / Gemini 3.1 Flash-Lite Image (default, all plans), Nano Banana 2.1 (`nano-banana-2`), Nano Banana Pro / Gemini 3 Pro Image. Every Google account, including a free one, gets 50 Flow credits a day, enough for 2 Veo 3.1 Fast or 5 Veo 3.1 Lite videos, and images cost no credits; Omni 1.1 Flash and video upscaling need a paid subscription. **Asset management**: list the projects on an account, read one project's media (uploads flagged as `likelyUpload`), and permanently delete media by id, up to 100 per call. **Agent**: chat with Google Flow Agent, Flow's built-in creative assistant (`/agent/chats`): messages with up to 7 image refs, 1 video, 5 voices and 7 characters with `@`-mentions, replies streamed over SSE (`/agent/chats/{chatId}/stream`, resumable with `Last-Event-ID`), approve or reject spending (`ask` / `auto` approval mode), read, list, rename and delete chats, and set the project's instructions and default image and video models (`/agent/settings`, which also lists the models the account's plan has with their credits). Chatting costs no credits, generated media cost the usual Flow price ([overview](https://useapi.net/docs/api-google-flow-v1/agent), [live demo](https://useapi.net/demos/google-flow/) with [the API call behind each feature](https://useapi.net/docs/api-google-flow-v1/agent-demo)). **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/google-flow/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 12 tools to chat with Google Flow Agent, answer its approvals, list, read, rename and delete chats, read and change the agent settings, list the account's characters, voices, images and videos, and upload images, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-google-flow-v1/mcp)). Includes captcha-solver pipeline and IPv6 rotation for resilience. See [Model Matrix](https://useapi.net/model-matrix) for current availability.

## Google Vids API

Runs on your [Google AI subscription](https://gemini.google/subscriptions/): Free, Plus $4.99, Pro $19.99, Ultra $99.99 or Ultra $199.99 a month (as of October 2026).

- [Google Vids API documentation](https://useapi.net/assets/aibot/api-google-vids-v1.txt): Third-party API for [Google Vids](https://docs.google.com/videos), Google's video editor with [Gemini Omni 1.1 Flash](https://blog.google/products-and-platforms/products/workspace/gemini-omni-in-google-vids/). **Video**: text-to-video, image-to-video from a start image, up to 3 reference images or avatars (`@referenceImage_1..3`, `@avatar_1..3` prompt mentions), 3-10s in 720p or 1080p, landscape or portrait. Extend (charged only for the added seconds), upscale to 1080p, and edit of clips made by the API. **Images**: one JPEG per call, 16:9 / 1:1 / 9:16 at about 1 MP, style presets. **Avatars**: an image (from a description or an image made with the API) plus one of 30 voices (`GET /voices`, no token). Image uploads, media download, sync or async jobs with `replyUrl` webhooks. No captcha. Uses the account's monthly Vids allowance (video seconds shared by a Google family plan, images per account), separate from Google Flow credits — Google AI Ultra $199.99 gets 10,000 video seconds and 1,000 images a month. Free Google accounts get no video or image allowance and can only make avatars. Every MP4 carries Google's visible Gemini watermark. **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/google-vids/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 16 tools for videos (create, extend, upscale, edit), images, uploads, avatars, voices, jobs and accounts, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-google-vids-v1/mcp)).

## Muse API

Runs on your [Muse subscription](https://www.meta.com/help/subscriptions/1021145227643680/): Free, Power $20 or Maximum $100 a month (as of October 2026).

- [Muse API documentation](https://useapi.net/assets/aibot/api-muse-v1.txt): Third-party API for [Muse](https://muse.ai), Meta's personal AI agent, on your own Muse account (added with its email and password, session renewed by the API). **Images** (`POST /jobs/image`): text to image or image to image from up to 9 input images (`image_1..9`): combine people, objects and backgrounds, edit an outfit, restore and colorize old photos. WebP of about 2-2.5 MP. **Videos** (`POST /jobs/video`): about 10 s at about 720p with AAC audio and speech from quoted lines, text to video, image to video, video to video (e.g. replace the person in a clip from a photo), inputs up to 9 images, 3 videos (`video_1..3`) and 3 audio files (`audio_1..3`). No model, aspect-ratio, resolution or duration parameters: say the aspect ratio in the prompt ("in exactly 16:9 aspect ratio"). Inputs are asset ids from `POST /assets` (raw image/video/audio body or JSON `{url}` of a public https file, up to 50 MB, stored in the account's Muse workspace) or from a previous job's `result.assets[]`, so generated files chain without re-upload. **Library**: `GET /assets` lists the account's whole Muse Library newest first (`type` image, video, audio, document, web, podcast; media also by origin: uploaded, generated, made), `GET /assets/{assetId}` streams any item (a document also as PDF with `?format=pdf`, a web app as one HTML file, a podcast as MP3 and its cover with `?part=cover`), `DELETE /assets/{assetId}` removes it, `POST /assets/{assetId}/share` publishes a web app the agent built on a public link (Muse reviews it first, 422 with the issues when refused) and `DELETE /assets/{assetId}/share` stops sharing. **Workspace files**: `GET /files` (list a folder), `GET /files/content` (download any file), `GET /files/search` (by name or extension), `POST /files` (write any file up to 50 MB) and `DELETE /files` cover the whole account workspace, including the code, documents and data the agent writes; chat `files` may name workspace paths (`file_1`…). **Chat** (`POST /chats`, `GET /chats/{chatId}/stream`): a two-way conversation with the whole Muse agent (web research, code, connected apps, documents, web apps, podcasts, images and videos) streamed over Server-Sent Events: text as it is written, files as asset ids (media and documents), the agent's emoji `reaction` events, `browser` events for its own web browser, `turn.done` per answer, with emoji reactions on messages via `POST /chats/{chatId}/messages/{messageId}/reactions` (the same emoji again removes it) and document asset ids accepted in chat `files`; send more any time with `POST /chats/{chatId}/messages` (also mid-answer), stop with `/cancel`, answer the agent's permission requests (`approval` events: allow once, always allow the site, deny) with `POST /chats/{chatId}/approvals/{approvalId}`, resume with `Last-Event-ID` (exactly where it left off); agent messages can carry widgets (`option` quick replies answered with `POST /chats/{chatId}/widgets/{widgetId}`, `html` cards to show in a sandboxed iframe, `space_build_completed` for a web app the agent built with its asset id, other kinds as `data` with `fallbackText`); about 50 open streams per account, approximate, listed by `GET /chats/streams`; chat takes no `maxJobs` slot (jobs only); list, read and delete chats with `GET /chats`, `GET /chats/{chatId}` (messages with files, widgets and reactions, plus `browsers`), `DELETE /chats/{chatId}` (the account's main chat answers 409). **Live browser** ([guide](https://useapi.net/docs/api-muse-v1/live-browser-guide)): watch the agent's own web browser in a chat and take control of it: `GET /chats/{chatId}/browser/{taskId}` (state, control, display), `POST /chats/{chatId}/browser/{taskId}/viewer` (a single-use live view link at `https://api.useapi.net/v1/muse/viewer#t=…`, `mode` view or control, opened within `ttl` 15-300 s, sessions up to 30 minutes, one viewer per account), `DELETE …/viewer` (revoke every viewer), `POST …/release` (give control back to the agent), `POST …/stop` (close the browser; at most 2 browsers run per account). **Account reset**: `POST /accounts/{email}/reset` with `{"confirm": "<email>"}` gives the account a fresh agent computer and permanently deletes every chat, file and memory on it. Sync (waits up to ~100 s, then 202) or async (201) jobs with `replyUrl` webhooks, `GET /jobs/{jobid}?history=true` returns the job's Muse chat for troubleshooting. **Agent permissions** per account: `GET /accounts/{email}/approvals` (pending and decided permission requests across chats), `GET`/`POST /accounts/{email}/permissions` (whether the agent asks before acting on websites and connected apps: `auto_allow` or `always_ask`), and change or delete standing allow/deny rules with `/accounts/{email}/permissions/{permissionId}`. Weekly Muse allowance per account (free plan works: about 2-3% per video, 0-1% per image), live in `GET /accounts/{email}`. **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/muse/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 37 tools for images, videos, chat, approvals, the agent's browser, the Library, workspace files, permissions and account reset, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-muse-v1/mcp)).

## Gemini Notebook API (NotebookLM API)

Runs on your [Google AI subscription](https://gemini.google/subscriptions/): Free, Plus $4.99, Pro $19.99, Ultra $99.99 or Ultra $199.99 a month (as of October 2026).

- [Gemini Notebook API (NotebookLM API) documentation](https://useapi.net/assets/aibot/api-gemini-notebook-v1.txt): Third-party API for Google's [Gemini Notebook](https://notebook.google.com) (formerly NotebookLM). **Notebooks**: create, list, rename, delete, copy (with sources and artifacts), guide (summary, suggested reports and questions). **Sources**: web pages and YouTube videos by URL, pasted text, Google Drive files and Google Docs (with Drive sync), raw-body file upload (PDF, Word, PowerPoint, EPUB, Markdown, text, CSV, audio, video, image), ebooks from the account's Google Play Books library, source guide (summary, key topics) and indexed text, passage search across sources, append text to a source, copy sources to another notebook, rename, delete. **Research**: Discover sources (web or the account's Google Drive) and Deep Research (report in Markdown) as jobs, then import the chosen sources and the report. **Notes**: create, list, edit, delete, convert to a source, save a whole chat conversation, mind map note (JSON tree, one call). **Source labels**: auto-label by topic (all or unlabeled only), named labels with emoji, add or remove sources. **Collections**: group notebooks into named folders with emoji. **Sharing**: add or remove viewers and editors, public or restricted link, allow copies, view level (full or chat only, paid plans). **Chat history and settings**: read or delete a conversation, goal (default, learning guide, custom persona) and answer length. **Account output language** for Studio artifacts. **Chat**: grounded answers with numbered citations and follow-up conversations. **Studio artifacts** (async jobs with `replyUrl` webhooks, sync or async): Audio Overview (deep dive, brief, critique, debate), Video Overview (explainer, brief, short, cinematic, with styles), slide deck (with one-slide revise), infographic, report (briefing, study guide, blog post, custom), interactive report (with instructions, and its suggested infographics, slide decks, mind maps, quizzes and flashcards generated one by one), data table, quiz, flashcards, mind map. Files (m4a, mp4, pdf, pptx, png) come from a download endpoint. Reports export to Google Docs and tables to Google Sheets, and finished artifacts copy to another notebook. One-shot mode generates straight from sources. Live per-account usage (5-hour and weekly windows) and quota-aware account selection. Free Google accounts make everything except cinematic Video Overviews (Pro and Ultra), and paid Google AI plans raise the limits. **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/gemini-notebook/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 57 tools for notebooks, sources (including file upload and Play Books), chat, notes, labels, sharing, collections, Studio artifacts, jobs and research, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-gemini-notebook-v1/mcp)).

## Dreamina API

Runs on your [Dreamina subscription](https://dreamina.capcut.com/pricing/dreamina-price): Free, Basic $15, Standard $36, Advanced from $79 or Ultra $520 a month in the US (as of October 2026).

- [Dreamina API documentation](https://useapi.net/assets/aibot/api-dreamina-v1.txt): Third-party API for [Dreamina](https://dreamina.capcut.com) by ByteDance. **Video**: Seedance 2.5 (480p/720p/1080p, 4-30s or 31-180s in long video mode, video edit via editVideoRef, omni up to 50 refs: 30 images, 10 videos, 10 audio clips, accepts real faces), Seedance 2.0 (720p/1080p/4K, 4-15s, omni: image+video+audio, keyframe + multi-ref up to 12 assets: 9 images, 3 videos, 3 audio clips), Seedance 2.0 Fast and Seedance 2.0 Mini (720p, plus 1080p/2K/4K upscaled from 720p), Seedance 1.5 Pro (720p, 1080p on CA), Seedance 1.0 Pro (1080p, CA only), Seedance 1.0 Fast (720p/1080p, 2-10 keyframes, CA only), HappyHorse 1.1 (4-15s, 720p/1080p, 9 image refs, CA only), Wan 3.0 (4-15s, 720p/1080p, first/last frame, 10 images + 5 videos + 5 audio refs, CA only), Seedance 1.0 Mini (720p, US only, 2-10 keyframes). **Images**: Seedream 5.0 Pro (CA, 1.5K/2K/4K), Seedream 5.0 Lite, Seedream 4.7 (CA), Seedream 4.6 (default), Seedream 4.5, Seedream 4.1, Seedream 4.0 (4K free on Advanced and Ultra until November 6, 2026), Seedream 3.0, Nano Banana / Gemini 2.5 Flash (CA region only, 1K, 3 refs), Nano Banana Pro (1K/2K/4K), GPT Image 2 (1K/2K/4K), GPT Image 2.5 Flare and Sunburst (1K/2K/4K). **Other**: video upscale 2x (720p to 1440p, 1080p to 4K), 30 or 60 FPS interpolation, image upscale to 2K/4K/8K. A free Dreamina account makes Seedream 4.6 images on Dreamina's free daily credits; Seedance video needs a paid subscription. 1080p, 4K and the CA-only models need a Canada-region account. See [Model Matrix](https://useapi.net/model-matrix) for cross-API availability.

## Kling API

Runs on your [Kling AI subscription](https://kling.ai/app/membership/membership-plan): Free, Standard $8.80, Pro $32.56 or Premier from $80.96 a month, with up to 26% off the first month (as of October 2026).

- [Kling API documentation](https://useapi.net/assets/aibot/api-kling-v1.txt): Third-party API for [Kling AI](https://kling.ai/app) by Kuaishou Technology for prompt adherence and consistency. **Video models**: Kling 4.0 (3-30s, stereo audio always on, 720p, or 1080p on paid plans, 21:9, multi-shot, keyframes between start and end frames (10 images in total, at most duration in seconds + 1), omni with up to 10 images, 7 elements incl. VIDEO elements, up to 5 reference videos of 3-30s each (30s total) with selectable output duration, and up to 3 voices, including your own voices saved from a 4.5-30s audio clip), Kling v3 / O3 (4K, audio, elements, motion control, omni), Kling v3 Turbo (faster v3, text-to-video + image-to-video), Kling O1 (elements), Kling 2.6 (native audio + motion control), Kling 2.5, 2.1, 1.6, 1.5. **Audio + lip-sync**: Avatars 2.0 (lip-sync video with TTS), Kling Audio (video audio generation), Motion Control (apply motion from video to image, v3.0 / v2.6). **Other**: Elements (reusable character/object references), Custom Voices (clone from MP4 video), Image Enhance (improve quality and restore details of any image, 2k or 4k setting), Text-to-speech (up to 5 minutes). Text-to-speech and face detection cost 0 credits on every Kling AI plan, the free one included. See [Model Matrix](https://useapi.net/model-matrix). **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/kling/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 45 tools for text / image / omni / motion / avatar video, lip-sync, sound, extend, effects, images, try-on, elements, voices, avatars, uploads and tasks, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-kling-v1/mcp)).

## PixVerse API

Runs on your [PixVerse subscription](https://app.pixverse.ai/subscribe): Basic (free), Standard $10, Pro $30, Premium $60, Ultra $199 or Team Ultra $199 a seat a month (as of October 2026).

- [PixVerse API documentation](https://useapi.net/assets/aibot/api-pixverse-v2.txt): Third-party API for [PixVerse](https://pixverse.ai) by AIsphere. **Native video**: PixVerse v6, v5.6, v5.5, v5, v5-fast, PixVerse C1. **Third-party video** (via PixVerse): Seedance 2.5 (480p/720p/1080p, up to 30s, native audio, omni up to 50 refs, Pro+ plans only), Seedance 2.0 / 2.0 Fast / 2.0 Mini (omni: image + video + audio references), Kling O3 / V3, Veo 3.1 Lite / Standard / Fast, Grok Imagine, Grok Imagine 1.5, HappyHorse 1.0, MiniMax H3 (Hailuo 3.0 — 768p/2K, 5-15s), MiniMax H3 Max (480p/768p/1080p, 5-15s), Gemini Omni Flash (720p, 3-10s, image + video references), FLUX 3 (Black Forest Labs — 720p/1080p, 5-20s, text-to-video and image-to-video), Wan 3.0 (Alibaba — 480p/720p/1080p, 2-30s, all four modes, 10 images + 5 video + 5 audio references). **Images**: Qwen Image, Nano Banana / Nano Banana 2 / Nano Banana 2.1 (up to 4K, 14 refs, 1:8 to 8:1 ratios, minimal/medium/high thinking, Standard+ plans) / Nano Banana 2 Lite (1080p, up to 14 refs, free on Pro+) / Nano Banana Pro, GPT Image 2.0 (up to 4K, 9 refs), GPT Image 2.5 Flare / 2.5 Sunburst (up to 4K, 16 refs, five detail levels through max, +5 credits per reference, Standard+ plans), Seedream 5.0 Pro (up to 1440p, 10 refs) / Seedream 5.0 Lite, Seedream 4.5 / 4.0, Kling O3 / 3.0. **Music**: music-2.6 (MiniMax, default), music-v1 (ElevenLabs), Lyria 3 Pro (Google) — songs, instrumentals, custom or model-written lyrics. **Speech**: speech-2.8-hd / speech-2.8-turbo (MiniMax, emotion control), eleven-multilingual-v2 / eleven-v3 (inline audio tags) / eleven-turbo-v2.5 (ElevenLabs) — 300+ voices, per-voice settings, 30 to 40 languages per model. **Other**: Upscale for your own videos and images (PixVerse Upscale, Topaz Proteus / Starlight Fast / Starlight Precise / Gigapixel, up to 4K, free price quote), Extend, Lip Sync, Modify, 900+ video effects. Relax Mode images cost 0 credits: Qwen Image and Nano Banana 2 Lite on Pro, plus Nano Banana, Nano Banana 2, Nano Banana 2.1 and Seedream 4.0 on Premium, and all 15 image models on Ultra; past a few hundred images per model a day, PixVerse moves the account to a slow queue. A free Basic account gets 60 credits to start and 30 a day for the PixVerse models up to 540p, and every third-party video model needs Pro or higher. See [Model Matrix](https://useapi.net/model-matrix).

## Mureka API

Runs on your [Mureka subscription](https://www.mureka.ai/subscribe): Free, Pro $10 or Premier $30 a month (as of October 2026).

- [Mureka API documentation](https://useapi.net/assets/aibot/api-mureka-v1.txt): Third-party API for [Mureka AI](https://www.mureka.ai) by Kunlun Tech, positioned to compete with Suno and Udio. **Music**: Mureka models — flagship V9.5 (default), V9 and O2, plus V8, V7.6 and V7.5-all — generate songs from lyrics/descriptions/musical styles, with vocal references, motif (melody) seeding, custom vocal cloning, reusable characters (your own voice as a singer, combined with any style), remix of a song into a new style, sheet music (MusicXML and PDF) from a song or YouTube link, extend, section editing over an existing song, and stem extraction (up to 12 instrument and vocal tracks plus MIDI, or vocals + instrumental, with DAW sessions for Ableton Live, Logic Pro, Pro Tools and Cubase). **Speech**: text-to-speech with Mureka's voices or your cloned voice (cloning needs Premier), with multi-speaker conversations. A free Mureka account makes 4 songs a day on V7.5-all, 2 at a time; the other models need Pro or Premier. See [Model Matrix](https://useapi.net/model-matrix). **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/mureka/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 40 tools for songs (prompt, lyrics, instrumental), characters, edits, remix, stems, sheet music, uploads, speech and voice cloning, jobs and accounts, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-mureka-v1/mcp)).

## Flow Music API

Runs on your [Flow Music subscription](https://www.flowmusic.app/pricing): Free, Starter $8, Plus $24 or Member $64 a month (as of October 2026).

- [Flow Music API documentation](https://useapi.net/assets/aibot/api-flowmusic-v1.txt): Third-party API for Google's [Flow Music](https://www.flowmusic.app), powered by the [Lyria](https://deepmind.google/models/lyria/) models — `lyria-3.5` (flagship) and `lyria-3-pro` (default), selectable per request via `model`. **Music**: generate complete songs from a text prompt (vocal or instrumental, optional custom lyrics + reference images/audio); edit clips via `cover` (restyle), `remix` (lyrics-adjust, deterministic UUID5 lyrics_id), `extend`, `replace`, audio `effect` (Trim/Fade/Gain/Reverb/Pan/Equalizer) and four stems (drums, bass, vocals, other); lyrics-only generation (renders no song, but uses a small amount of credit); AI image generation. Sync by default, opt-in `async` with `replyUrl` webhooks. Sign in with a Google account at flowmusic.app. Each song uses 5 credits of the Flow Music subscription, a free account makes songs from its daily top-up credits, and a Google AI subscription doesn't include Flow Music credits. Flow Music was formerly Producer AI (originally Riffusion). See [Model Matrix](https://useapi.net/model-matrix) for cross-API availability. **MCP server**: a hosted Model Context Protocol server at `https://api.useapi.net/v1/flowmusic/mcp` (Streamable HTTP, `Authorization: Bearer <API token>`) with 18 tools for songs, lyrics, extend, replace, cover, lyric remix, effects, stems, reference uploads, jobs and accounts, for Claude Code, Cursor, VS Code and other MCP clients ([setup](https://useapi.net/docs/api-flowmusic-v1/mcp)).

## MiniMax API

Video and images run on your [Hailuo AI subscription](https://hailuoai.video/subscribe): Free, Standard $10.50, Pro $38, Master $89 or Max $216 a month. Songs and speech run on a separate [MiniMax Audio subscription](https://www.minimax.io/audio/subscribe): Free, Starter $5, Creator $15, Standard $30 or Pro $99 a month (as of October 2026).

- [MiniMax API documentation](https://useapi.net/assets/aibot/api-minimax-v1.txt): Third-party API for [MiniMax](https://www.minimax.io) through the [Hailuo AI](https://hailuoai.video) website and MiniMax Audio. **Video**: Hailuo 3.0 (MiniMax H3, 4-15s at 768p / 2K, sound on every video, references up to 9 images, 3 videos and 3 audio clips), Hailuo 2.3 / 2.3 Fast and Hailuo 02 (unlimited on Max), Hailuo 01 (unlimited on Max), Veo 3.1 / 3.1 Fast (720p, 1080p or 4K), Seedance 2.0 (up to 4K) / Fast / Mini (omni: image + video + audio references). **Images**: Midjourney V7, Midjourney Niji 7, Nano Banana Pro and Nano Banana 2 (up to 4K, 14 refs), GPT Image 2 (up to 4K, 16 refs) / GPT Image 1.5, Seedream 5.0 / 4.5, image-01. **Music** (MiniMax Audio subscription): Music 3.0 / Music 2.6 — songs from a text prompt, your own or model-written lyrics, instrumentals, covers of an uploaded MP3/WAV reference or of a previously generated song, up to 3 tracks per call, with optional real-time audio streaming. **Speech** (MiniMax Audio subscription): ten text-to-speech models (Speech 2.8/2.6/2.5/02/01, each in hd and turbo) via [POST speech/create](https://useapi.net/docs/api-minimax-v1/post-minimax-speech-create) — 600+ voices in 40 languages, 5,000 characters a call or 10,000 on turbo, emotion/sound/pause markup, word timings, voice cloning, replyUrl callback or SSE streaming. A free MiniMax Audio account gets 10,000 credits a month (about 33 songs), and a free Hailuo AI account gets one-time trial credits for Hailuo models only. Multi-account load balancing included. See [Model Matrix](https://useapi.net/model-matrix).

## Runway API

Runs on your [Runway subscription](https://runway.com/pricing): Free, Standard $15, Pro $35, Max $95 or Team $69 a seat a month (as of October 2026).

- [Runway API documentation](https://useapi.net/assets/aibot/api-runwayml-v1.txt): Third-party API for [Runway](https://runway.com) by Runway AI. **Native video**: Gen-4.5, Gen-4, Gen-4 Turbo, Gen-4 Aleph (video-to-video). **Third-party video** (via Runway): Kling O3 4K / Pro / Standard (5-15s with sound, up to 4K, 2 to 6 shots, video edit on Pro and Standard), Kling v3 / 2.6, Veo 3.1, Wan 3.0 / Wan 3.0 Prime (2-30s, up to 1080p, 10 image + 5 video + 5 audio references), MiniMax H3 (Hailuo 3, 768p/2K, 9 image + 3 video + 3 audio references) / MiniMax H3 Max (480p/768p, first/last frame, prompt expansion), FLUX 3 (5-20s, 720p/1080p, first frame, video extend), Wan 2.6 Flash / 2.2 Animate, HappyHorse 1.0, Grok Imagine 1.5, Seedance 2.5 (480p/720p/1080p, up to 30s, 50 references, edit and extend video), Seedance 2.0 / 2.0 Fast / 2.0 Mini. **Images**: FLUX 3 Image (up to 4K, 10 refs), FLUX.2 Max (up to 4K, 8 refs) / FLUX.2 Klein (up to 2K, 4 refs), Nano Banana / Nano Banana 2 / Nano Banana 2 Lite / Nano Banana Pro, GPT Image 1 Mini / 1.5 / 2.0 (up to 4K, 14 refs), Seedream 5.0, Gen-4 / Gen-4 Turbo. **Other**: Act Two (character animation), Lip Sync, Frames, Video Upscale (Topaz AI 4K, any source video, up to 40s), Video Extend, Kling Motion Control. Image descriptions cost 0 credits on every Runway account, the free one included, and Runway's Free plan comes with 125 one-time credits. **Runway Unlimited plan** unlocks credit-free `exploreMode` on most image and video models. See [Model Matrix](https://useapi.net/model-matrix).

## TemPolor API

Runs on your [TemPolor subscription](https://www.tempolor.com/pricing): Lite $9.79, Pro $13.99 or Ultra $29.99 a month, with no free plan (as of October 2026).

- [TemPolor API documentation](https://useapi.net/assets/aibot/api-tempolor-v1.txt): Third-party API for [TemPolor](https://www.tempolor.com) by Qustar Technology, a royalty-free music generation service. **Models**: TemPolor v4.8 (`v4.8`, songs with vocals from a prompt or your lyrics, up to 5 minutes in 30+ languages, sung by a cloned voice or an artist voice, or built on your MIDI file) and TemPolor i4 (`i4`, instrumentals with the key, tempo and chords you set); older model names run on v4.8 or i4. Stem splitter for your own audio, and every track downloads as mp3, wav or stems. Lite and Pro include 1,000 and 3,000 tracks a month, and the **Ultra plan** has no per-song charge and runs 10 generations at once. See [Model Matrix](https://useapi.net/model-matrix).

## InsightFaceSwap API

Runs on your [InsightFaceSwap subscription](https://www.patreon.com/picsi/membership) on Patreon: Free, Basic $9.99, Pro $19.99 or Ultra $34.99 a month (as of October 2026).

- [InsightFaceSwap API documentation](https://useapi.net/assets/aibot/api-faceswap-v1.txt): Third-party API for the [InsightFaceSwap](https://discord.gg/Ym3X8U59ZN) Discord Bot by [Picsi.Ai](https://www.picsi.ai). Face swap from source images onto target images. Features: HiFidelity Mode, ARTIFY effects, age transformation, multi-face morphing, background change, headshot generation. The free plan works with 10 credits a day (it swaps the largest face in a static image), and paid plans add GIF swaps, multi-face swaps and the paid modes. See [Model Matrix](https://useapi.net/model-matrix).

## Optional

Secondary material an agent can skip when context is tight.

- [General Q&A](https://useapi.net/assets/aibot/qa.txt): Cross-cutting tips (rate limiting, prompt moderation, common patterns). **WARNING**: items are general-purpose and may reference deprecated APIs, sunsetted services, or scenarios not applicable to a specific API — always cross-check against the relevant service-specific documentation before applying.
- [Legal & Terms of Service](https://useapi.net/docs/legal): Terms of service and legal information for the useapi.net third-party APIs.
- [All blog posts](https://useapi.net/assets/aibot/blog.txt): every announcement in date order, combined.