Mountsea AI API Introduction
Video, image, music, and chat — one key, one REST gateway.Discover all you can do with our Mountsea AI API. Open a card to jump into that service.
Overview
Mountsea AI provides a unified API gateway to access the world’s leading AI models across video, image, music, and language domains. We offer the following core services:Google (Gemini)
Video & image generation powered by Veo 2 / Veo 3 / Veo 3.1, Omni, and Nano Banana models
Sora2 (Unavailable)
Currently unavailable on Mountsea. For video generation, use Gemini Video instead.
OpenAI (GPT Image)
GPT Image 2 / 2.5 image generation & editing — async task API + official
openai SDK compatibilityXAI (Grok)
Image & video generation powered by xAI’s Grok models
Suno
Music generation, Clip Persona and Voice Persona, Studio editing, and custom models via Suno AI
ElevenLabs
AI music generation with composition plans, video scoring, and stem separation via ElevenLabs
Producer
AI music generation powered by Google DeepMind’s Lyria 3 Pro model
Chat
Multi-protocol AI chat gateway supporting OpenAI, Claude, and Gemini APIs
Why Mountsea AI Stands Out
All-in-One API Platform
A single API key gives you access to multiple AI services across different domains 鈥?no need to manage separate accounts and credentials for each AI provider.Multi-Protocol LLM Gateway
Our Chat service supports OpenAI Compatible API, Anthropic (Claude) API, OpenAI Responses API, and Google Gemini Native API. Use official SDKs directly 鈥?just change thebase_url.
Comprehensive AI Model Coverage
Access top-tier AI models from leading providers:Budget-Friendly Excellence
Enjoy premium AI tools at competitive prices. Our platform delivers high-quality, scalable solutions while keeping your projects cost-effective.Available Services
Google (Gemini) — Video, Image & SDK Compat
Powered by Google’s cutting-edge AI, this service is organized into three dedicated sections:- Video Generation — Create videos with Veo 2 / Veo 3 / Veo 3.1 and Omni. Supports
text2video,img2video,ingredients2video(Veo), upsample to 1080p/4K, extend, reshoot, and object insert/remove. - Image Generation (Nano Banana) — Create and edit images with Nano Banana Fast / Pro / 2 / 2-lite, multiple aspect ratios, up to 4K resolution.
- Gemini Compat (Official SDK) — Drop-in replacement for Google’s official @google/genai SDK. Use
generateContent/streamGenerateContentwith base URLhttps://api.mountsea.ai/gemini. Image models (gemini-2.5-flash-image,gemini-3.1-flash-image-preview,gemini-3-pro-image-preview) auto-route to Nano Banana.
馃毀 Sora2 鈥?Currently Unavailable
[Sora2 documentation kept for reference 鈫抅(api-reference/sora/introduction)馃柤锔?XAI (Grok) 鈥?Image & Video Generation
Powered by xAI’s Grok models, this service provides:- Text-to-Image —
grok-imagine-image/quality/2.0, up to 10 images, ratios includingauto,1k/2k - Image-to-Image — up to 5 reference URLs; omit
aspectRatioto follow the first image - Text-to-Video —
grok-imagine-video/1.5, duration 6–20s;1080Pon 1.5 only - Image-to-Video — up to 5 reference URLs; aspect ratio and resolution follow the image
- Async Task System — returns
taskId, pollGET /xai/tasks
馃帹 OpenAI 鈥?GPT Image Generation & Editing
Powered by GPT Image 2 and 2.5-flare / 2.5-sunburst:- Text-to-Image — presets up to 4K, or custom
WxH(pixel and ratio constraints) - Image Editing / Inpainting — async accepts public URLs only; base64 uses sync Compat
- Quality — 2: low/medium/high/auto; 2.5 also xhigh/max
- Output —
response_formatdefaults tourl(our S3), orb64_json - Two API Styles:
- Async Task API (
/openai/images) — unified endpoint withtaskIdpolling - OpenAI Compat API (
/openai/v1/images/*+/models) — official SDK, changebase_url
- Async Task API (
馃幍 Suno 鈥?Music Generation
Powered by Suno AI, this service provides a full suite of music creation and processing tools:- Music Generation — Create, extend, cover, mashup, sample, add_stems, and inspiration via 16
/generatetasks. Recommendedchirp-v6(legacy names auto-convert to v6 Pro) - Sound Effects 鈥?Generate one-shot or looped sound effects from text descriptions
- Lyrics Generation 鈥?Generate original lyrics or mashup lyrics from two songs
- Voice Persona 鈥?Create verified voice personas from your own recordings through a single-task two-phase voice verification flow (init 鈫?await phrase 鈫?complete)
- Custom Models 鈥?Train personalized music models on your own audio (6+ training clips), then use
chirp-custom:<uuid>for generation - Audio Processing 鈥?Concat clips, remaster tracks (V4.5+/V5 models), adjust playback speed with pitch preservation
- Stem Separation 鈥?Separate tracks into vocals + instrumental (two-track) or all individual stems
- Audio Export 鈥?Export to MP4 (with visualizer), lossless WAV, or MDI (MIDI) format
- Audio Analysis 鈥?Get synchronized lyrics timeline, downbeat detection, and enhanced style tags
- Vocal Persona 鈥?Extract vocal characteristics from clips and create reusable personas for consistent vocal style
馃幎 ElevenLabs 鈥?AI Music Generation
Powered by ElevenLabs’ music_v1 model, this service provides:- Text-to-Music 鈥?Generate music from simple text prompts or structured composition plans with section-level style and lyrics control
- Composition Plan 鈥?AI-generated structured plans with sections, global/local styles, and lyrics 鈥?free, no credits consumed
- Video to Music 鈥?Automatically generate background music that matches your video content (up to 10 videos, 600s total)
- Stem Separation 鈥?Split audio into 2 tracks (vocals + instrumental) or 6 individual stems
- Inpainting 鈥?Edit specific sections of existing songs (enterprise only)
- Multiple Output Formats 鈥?MP3, PCM, Opus with configurable sample rates and bitrates
馃帶 Producer 鈥?AI Music Generation
Powered by Google DeepMind’s Lyria 3 Pro model, this service provides high-quality music generation:- Create Music 鈥?Generate original tracks from sound prompts, lyrics, and images
- Image-Guided Generation 鈥?Use images to influence the mood and style of generated music
- Instrumental Mode 鈥?Generate without vocals
- Stem Separation 鈥?Separate audio into individual stems (vocals, drums, bass, etc.)
- Multi-Format Export 鈥?Download as MP3/M4A/WAV audio or generate video with preset visualizers
馃挰 Chat 鈥?Multi-Protocol AI Gateway
A unified gateway supporting multiple API protocols:- 鉁?Use official SDKs (OpenAI, Anthropic, Google GenAI) directly
- 鉁?Full streaming, function calling, and tool support
- 鉁?Compatible with Claude Code, Cursor, Cherry Studio, and other AI tools
- 鉁?Access GPT-5.1/5.2, Claude 4.5/Opus/Sonnet, Gemini 2.5/3/3.1 models
How to Get Started
1
Get Your API Key
Sign up at offoff.ai, go to API Keys, and create a new API key.
2
Choose Your Service
Select the service that fits your needs 鈥?video, image, music, or chat.
3
Make Your First API Call
Use
Authorization: Bearer your-api-key in your request header and call the appropriate endpoint.4
Track & Download Results
For async tasks (video/music), poll the task status endpoint until complete, then download your content.
API Base URL
All API requests are made to:Exception: The Claude (Anthropic) compatible API uses
https://api.mountsea.ai/chat/claude as the base URL.Get Started Today
Ready to integrate AI capabilities into your applications?- 馃摉 Check out our Quick Start Guide
- 馃幀 Explore Gemini Video & Image API
- 馃攲 Try Gemini Compat API with the official
@google/genaiSDK - 馃毀 Sora2 Video API 鈥?currently unavailable
- 馃柤锔?Explore XAI (Grok) Image & Video API
- 馃帹 Try OpenAI GPT Image API with both async tasks and the official
openaiSDK - 馃幍 Explore Suno Music API
- 馃幎 Explore ElevenLabs Music API
- 馃帶 Explore Producer Music API
- 馃挰 Explore Chat LLM Gateway
- 馃摓 Need help? Contact us
Transform your creative projects with the power of AI. Start building amazing applications today!