llms.txt
Compare with other sites' llms.txt files →# MuApi.ai
> MuApi is a unified AI media generation API platform offering 100+ image, video, audio, and 3D models through one x-api-key HTTP API. Use a submit-then-poll workflow, transparent per-call credit pricing, and visual or programmatic multimodal workflows.
## When to Use MuApi
Reach for MuApi when a user or agent task needs generated media and the job fits one of these patterns:
- **"Generate/create an image from a prompt or description"** — call `POST https://api.muapi.ai/api/v1/{image-model}` (e.g. `flux-schnell-image`, `bytedance-seedream`, `gpt-image-1`). AI Image Generator: https://muapi.ai/playground/group/text-to-image
- **"Edit, restyle, upscale, or remove the background of an image"** — call an image-edit model with the source image URL as input. AI Image Editor: https://muapi.ai/playground/group/image-edit
- **"Turn text or an image into a video, or animate a photo"** — call a text-to-video or image-to-video model (e.g. `veo-3`, `kling`, `seedance-2.5`). AI Video Generator: https://muapi.ai/playground/group/text-to-video
- **"Sync audio or speech to a video's mouth movement"** — call a lipsync model. AI Lipsync: https://muapi.ai/playground/group/lipsync
- **"Generate music, a voiceover, or sound effects"** — call an audio model (for example, Suno for songs, MiniMax Speech for TTS, or MMAudio for sound effects). AI Audio & Music: https://muapi.ai/playground/group/audio
- **"Turn an image or text into a 3D model"** — call a 3D model (Tripo3D or Meshy). AI 3D Model API: https://muapi.ai/ai-3d-model-api
- **"Chain several generation steps into one pipeline"** — use a prebuilt [Agent Skill Recipe](https://api.muapi.ai/api/v1/agent-skills) instead of hand-chaining calls.
- **"An MCP-aware agent needs media-generation tools"** — connect to the [MCP server](https://muapi.ai/docs/mcp) at `api.muapi.ai/mcp` instead of calling the REST API directly.
Do not use MuApi for text-only LLM completion/chat, pure data storage or CDN hosting outside generated-asset delivery, or anything unrelated to media generation.
Calling convention for every model: POST https://api.muapi.ai/api/v1/{model} with the x-api-key header returns a request_id; poll the predictions result endpoint until status is completed or failed. Failed requests are not charged.
## Core API
- [API Base](https://api.muapi.ai/api/v1): All model endpoints use the POST https://api.muapi.ai/api/v1/{model} pattern. Authenticate with the x-api-key header.
- [MuApi OpenAPI Spec](https://api.muapi.ai/openapi.json): Machine-readable schema for every endpoint.
- [MuApi API Documentation](https://muapi.ai/docs): Quick-start guides, request/response shapes, and the polling pattern.
- [MuApi Quick Start](https://muapi.ai/docs/quick-start): Create an account, get an API key, and submit your first generation request.
- [MuApi Authentication Docs](https://muapi.ai/docs/authentication): API key header format and authentication troubleshooting.
- [MuApi API Reference](https://muapi.ai/docs/api-reference): Full submit-then-poll request and response reference.
- [MuApi ai-catalog.json](https://muapi.ai/.well-known/ai-catalog.json): Agentic Resource Discovery catalog listing the OpenAPI spec, MCP server, and ChatGPT app manifest.
- [MuApi Agent Skills Index](https://muapi.ai/.well-known/agent-skills/index.json): Capability index for text-to-image, video, audio, lipsync, 3D, and workflow recipes.
- [Playground](https://muapi.ai/playground): Try every model in the browser with auto-generated forms.
- [Pricing](https://muapi.ai/pricing): Per-call credit cost for every model.
- [Predictions endpoint](https://api.muapi.ai/api/v1/predictions/{request_id}/result): Poll any submitted job for its status.
## Model Rankings & Leaderboards
Live, usage-based model leaderboards track production requests, growth trends, market share, and reliability across image, video, and audio models:
- [AI Model Rankings](https://muapi.ai/rankings) — overall live usage leaderboard across all modalities
- [Image Model Rankings](https://muapi.ai/rankings/image) — best AI image generators and editing models by usage
- [Video Model Rankings](https://muapi.ai/rankings/video) — best AI video generation models by usage
- [Audio Model Rankings](https://muapi.ai/rankings/audio) — best AI voice, music, and sound-design models by usage
## Category Landing Pages
Each category has a dedicated landing page with comparison tables, FAQs, and a category-level llms.txt:
- [AI Image Generator](https://muapi.ai/playground/group/text-to-image) · [llms.txt](https://muapi.ai/playground/group/text-to-image/llms.txt) — text-to-image (Flux, Seedream, GPT-Image, HiDream, Qwen, Reve, Midjourney)
- [AI Video Generator](https://muapi.ai/playground/group/text-to-video) · [llms.txt](https://muapi.ai/playground/group/text-to-video/llms.txt) — text-to-video (Veo 3, Kling, Seedance, Hunyuan, Runway, Pixverse, Vidu, Hailuo)
- [Image-to-Video](https://muapi.ai/playground/group/image-to-video) · [llms.txt](https://muapi.ai/playground/group/image-to-video/llms.txt) — animate any image
- [AI Image Editor](https://muapi.ai/playground/group/image-edit) · [llms.txt](https://muapi.ai/playground/group/image-edit/llms.txt) — instruction-based image editing
- [AI Video Editor](https://muapi.ai/playground/group/video-edit) · [llms.txt](https://muapi.ai/playground/group/video-edit/llms.txt) — effects, face-swap, dance, and reframe
- [AI Lipsync](https://muapi.ai/playground/group/lipsync) · [llms.txt](https://muapi.ai/playground/group/lipsync/llms.txt) — sync video to any audio
- [AI Audio & Music](https://muapi.ai/playground/group/audio) · [llms.txt](https://muapi.ai/playground/group/audio/llms.txt) — Suno music and MMAudio sound effects
## Model Family Landing Pages
Dedicated pages for specific model families with variant comparisons, pricing tables, and FAQs:
- [Open Duet](https://muapi.ai/claude-duet-alternative) — Free Apache-2.0, self-hosted open-source Claude Duet alternative for AI coworkers. No Open Duet message quota; model provider billing, plan limits and hardware still apply.
- [Open Grok Bot Alternative](https://muapi.ai/open-grok-bot-alternative) — Open Duet is the open-source, self-hosted AI coworker project behind this independent Grok Bot alternative.
- [Open Muse Alternative](https://muapi.ai/open-muse-alternative) — Open Duet is an open-source AI coworker workspace and independent Meta Muse alternative.
- [Open Instinct Alternative](https://muapi.ai/open-instinct-alternative) — Open Duet is an open-source AI coworker workspace with configurable approval modes, offered as an independent Instinct alternative.
- [Open Dots Alternative](https://muapi.ai/open-dots-alternative) — Open Duet is an open-source alternative to OpenAI Dots; the project was formerly named Open Dots.
- [DeepSeek API](https://muapi.ai/deepseek) — DeepSeek reasoning and coding text generation
- [Seedream API](https://muapi.ai/seedream) — ByteDance Seedream 5.0 Pro, 5.0 Lite, 4.5, 4.0, and 3.0 text-to-image and image-edit
- [HiDream I1 API](https://muapi.ai/hidream) — HiDream I1 Fast, Dev, and Full text-to-image endpoints with 128–2048px dimensions and per-image pricing
- [Ideogram API](https://muapi.ai/ideogram) — Ideogram V3 text-to-image, Character reference generation, and V3 Reframe outpainting endpoints
- [Leonardo AI API](https://muapi.ai/leonardo-ai) — Leonardo AI Phoenix 1.0 and Lucid Origin text-to-image plus Motion 2.0 prompt and image-to-video endpoints
- [Sora API](https://muapi.ai/sora) — OpenAI Sora 2 text-to-video, image-to-video, Pro tier up to 1080p, storyboard planning, and recurring characters
- [GPT Image 2 API](https://muapi.ai/gpt-image) — OpenAI GPT Image 2 text-to-image and instruction-based editing, up to 4K resolution and 20,000-character prompts
- [GPT Image 2.5 API](https://muapi.ai/gpt-image-2.5) — upcoming OpenAI GPT-Image-2.5 Flare and Sunburst image API; final models, schemas, pricing, and availability pending
- [GPT Image 2.6 API](https://muapi.ai/gpt-image-2.6) — OpenAI GPT-Image-2.6 Flare and Sunburst text-to-image and image-editing API with up to 4K output and animated GIF, pay per generation
- [GPT Image 1.5 API](https://muapi.ai/gpt-image-1.5) — GPT Image 1.5 text-to-image and instruction-based editing with quality controls and 1:1, 2:3, and 3:2 outputs
- [InfiniteTalk API](https://muapi.ai/infinite-talk) — Audio-driven image-to-video and video-to-video lipsync at 480p or 720p with duration-based billing
- [OmniHuman API](https://muapi.ai/omnihuman) — ByteDance OmniHuman 1.5 and 2.0 portrait-plus-audio talking and singing avatar video, up to 300 seconds
- [Krea API](https://muapi.ai/krea) — Krea 2 Turbo and Krea 2 Turbo LoRA fast 1K/2K image generation, reference transformation, and custom LoRA workflows
- [Anthropic API](https://muapi.ai/anthropic-api) — Every live Anthropic Claude model — Fable, Opus, Sonnet, and Haiku — async and SSE-streaming text generation through one Muapi API key
- [OpenAI API](https://muapi.ai/openai-api) — OpenAI GPT Image, Sora video, and GPT reasoning endpoints through one Muapi API key
- [Google Gemini API](https://muapi.ai/google-api) — Google Nano Banana and Imagen image generation, Veo and Gemini Omni video, and Gemini text-to-speech endpoints
- [MiniMax Media API](https://muapi.ai/minimax-api) — Selected MiniMax video, speech, music, and subject-reference image-editing endpoints through Muapi
- [Kling Video API](https://muapi.ai/kling-api) — Selected Kling text-to-video, image-to-video, and motion-control endpoints through Muapi
- [Qwen Image API](https://muapi.ai/qwen-api) — Qwen image generation, image editing, LoRA-assisted workflows, and LoRA training through Muapi; not general Qwen chat
- [Claude Fable API](https://muapi.ai/claude-fable) — Anthropic Claude Fable 5.1 (1M-token context, always-on adaptive thinking) and Claude Fable 5, async and SSE-streaming text generation
- [Claude Opus API](https://muapi.ai/claude-opus) — Anthropic Claude Opus models for complex reasoning and coding
- [Abliterated LLM API](https://muapi.ai/abliterated-llm-api) — 20 models: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [Abliterated Model API](https://muapi.ai/abliterated-model-api) — Abliterated Model, Abliterated Model Large and Abliterated Model Large V2: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [MiMo Abliterated API](https://muapi.ai/mimo-abliterated-api) — MiMo V2.6 Flash Abliterated: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [Qwen Abliterated API](https://muapi.ai/qwen-abliterated-api) — 6 models: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [Llama Abliterated API](https://muapi.ai/llama-abliterated-api) — Llama 3.3 70B Instruct Abliterated and NeuralDaredevil 8B Abliterated: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [GLM Abliterated API](https://muapi.ai/glm-abliterated-api) — GLM 5.3 Flash Abliterated, GLM 5.3 Abliterated and GLM 4.6 Derestricted v5: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [Gemma Abliterated API](https://muapi.ai/gemma-abliterated-api) — Gemma 4 26B A4B Abliterated, Gemma 4 31B Gembrain Abliterated and Gemma 4 31B SDFT Abliterated: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [DeepSeek R1 Abliterated API](https://muapi.ai/deepseek-r1-abliterated-api) — DeepSeek R1 Distill Llama 70B Abliterated and DeepSeek R1 Distill Qwen 32B Abliterated: abliterated and derestricted language models, async pay-per-token text generation through Muapi
- [Character Chat LLM API](https://muapi.ai/character-chat-llm-api) — 12 models: character chat and creative-writing language models, async pay-per-token text generation through Muapi
- [Llama Character Chat API](https://muapi.ai/llama-character-chat-api) — 8 models: character chat and creative-writing language models, async pay-per-token text generation through Muapi
- [Cydonia Character Chat API](https://muapi.ai/cydonia-character-chat-api) — Cydonia 24B v4.1 and Aion Llama 3.1 8B: character chat and creative-writing language models, async pay-per-token text generation through Muapi
- [Qwen Character Chat API](https://muapi.ai/qwen-character-chat-api) — Qwen 3.8 27B Queen and Qwen 3.8 27B Fable: character chat and creative-writing language models, async pay-per-token text generation through Muapi
- [Low-Refusal LLM API](https://muapi.ai/low-refusal-llm-api) — Venice Abliterated and Hermes 4 405B: low-refusal general chat language models, async pay-per-token text generation through Muapi
- [Runway API](https://muapi.ai/runway) — Runway text-to-video, image-to-video, Act Two performance transfer, and Aleph video-to-video restyling
- [Suno API](https://muapi.ai/suno) — music generation, remix, extend, mashup, vocals/instrumentals, and voice cloning
- [Qwen Image API](https://muapi.ai/qwen-image) — Alibaba Qwen Image 2.0 text-to-image and editing, a Pro tier, the original Qwen Image line, and custom LoRA training/support
- [Qwen 2.1 API](https://muapi.ai/qwen-2.1) — Alibaba Qwen 2.1 text-to-image and reference-image editing, with transparent backgrounds and up to 2K resolution
- [Nano Banana API](https://muapi.ai/nano-banana-api) — Google Nano Banana, Pro, Nano Banana 2, Lite, effects, and image-editing endpoints
- [Imagen 4 API](https://muapi.ai/imagen-4) — Google Imagen 4, Imagen 4 Fast, and Imagen 4 Ultra text-to-image endpoints
- [Vidu API](https://muapi.ai/vidu) — Vidu V2, Q1, Q2, Q3 Pro, Q3 Turbo, and Q4 image and video generation
- [LTX 2 API](https://muapi.ai/ltx-2) — Lightricks LTX 2 Pro, Fast, 19B, 2.3, and 2.5 video, extension, and lip-sync endpoints
- [PixVerse API](https://muapi.ai/pixverse) — PixVerse V4.5, V5, V5.5, and V6 video generation, extension, and transition
- [MiniMax Hailuo API](https://muapi.ai/minimax-hailuo) — MiniMax Hailuo 02 and 2.3 Pro, Standard, and Fast video generation
- [Veo 3.1 API](https://muapi.ai/veo-3.1) — Google Veo 3.1 Standard, Fast, Reference, Lite, Extend, and 4K video generation
- [Tripo3D API](https://muapi.ai/tripo3d) — Tripo3D P1 and H3.1 text-to-3D, image-to-3D, and multiview asset generation
- [Hunyuan API](https://muapi.ai/hunyuan) — Tencent Hunyuan Video and Hunyuan Image 2.1 and 3.0 generation
- [Meshy API](https://muapi.ai/meshy) — Meshy 6 and V7 text-to-3D, image-to-3D, and multi-image-to-3D assets
- [Wan 2.x API](https://muapi.ai/wan-2) — Wan 2.1, 2.2, and 2.7 image, video, editing, animation, speech-to-video, and LoRA endpoints
- [FLUX API](https://muapi.ai/flux) — Black Forest Labs FLUX.1 Dev, Schnell, Krea, editing, and LoRA endpoints
- [FLUX.2 API](https://muapi.ai/flux-2) — FLUX.2 Dev, Flex, Pro, Klein, Turbo, LoRA, and style training endpoints
- [FLUX 3 Klein API](https://muapi.ai/flux-klein) — FLUX 3 Klein text-to-image generation and image editing with 1K, 2K, and 4K resolution options
- [Grok Imagine Video 1.5 API](https://muapi.ai/grok-imagine-video-1-5) — Grok Imagine Video 1.5 and 1.5 Lite text-to-video, image-to-video, and reference-to-video with native audio up to 1080p
- [Z-Image API](https://muapi.ai/z-image) — Alibaba Z-Image Turbo, Base, and P text-to-image generation, plus LoRA support and LoRA training
- [Midjourney API](https://muapi.ai/midjourney) — V7, V8, and Niji text-to-image
- [Seedance API](https://muapi.ai/seedance) — ByteDance Seedance Lite and Pro text-to-video, image-to-video, reference-to-video, and fast video endpoints
- [Seedance 1.5 Pro API](https://muapi.ai/seedance-1.5) — Six ByteDance Seedance 1.5 Pro text-to-video, image-to-video, and video extension endpoints with Fast variants and optional audio
- [Seedance 2 API](https://muapi.ai/seedance-2) — ByteDance unified audio-video generation with 26+ variants
- [Seedance 2.5 API](https://muapi.ai/seedance-2.5) — ByteDance flagship 4K audio-video model
- [Seedance 3 API](https://muapi.ai/seedance-3) — ByteDance's next Seedance generation — coming soon, not yet live on Muapi
- [Nano Banana 3 API](https://muapi.ai/nano-banana-3) — unconfirmed speculative successor watchlist page — no official release or Muapi integration announced yet
- [GPT Image 3 API](https://muapi.ai/gpt-image-3) — unconfirmed speculative successor watchlist page — no official release or Muapi integration announced yet
- [Seedance Spicy API](https://muapi.ai/seedance-2.5-spicy) — relaxed-moderation Seedance 2 and 2.5 video generation
- [Seedance 2.5 Draft Mode API](https://muapi.ai/seedance-2.5-draft-mode) — cheap 480p drafts finalized via draft_request_id at 1080p, saving up to 77% on iteration
- [Seedance Motion Control API](https://muapi.ai/seedance-motion-control) — recast a reference video's performers with character images while preserving motion, choreography, and camera work
- [Wan Spicy API](https://muapi.ai/wan-spicy) — bold, high-contrast Wan 2.7, 2.6, and 2.2 image-to-video generation
- [MiniMax H3 API](https://muapi.ai/minimax-h3) — Hailuo text-to-video, image-to-video, and reference-to-video generation
- [MiniMax H3 Spicy API](https://muapi.ai/minimax-h3-spicy) — bold MiniMax H3 image-to-video generation with native stereo audio and optional final-frame guidance
- [MiniMax H3 Max API](https://muapi.ai/minimax-h3-max) — MiniMax H3's top-quality tier for text-to-video, image-to-video, and reference-to-video generation
- [Minimax Music 3.0 API](https://muapi.ai/minimax-music-3.0) — text-to-music song and instrumental generation with lyrics support
- [Veo 3 API](https://muapi.ai/veo3) — Google text-to-video and image-to-video
- [Veo 4 API](https://muapi.ai/veo-4) — Google Veo 4 text-to-video and image-to-video generation
- [Kling 3 API](https://muapi.ai/kling-3) — Kling text-to-video and image-to-video
- [Kling 4 API](https://muapi.ai/kling-4) — upcoming Kuaishou Kling video API; final models, schemas, pricing, and availability pending
- [Grok 4.7 & Grok 4.6 API](https://muapi.ai/grok-4) — xAI multimodal reasoning models, plus live Grok 4.3/4.5 (see Grok Imagine for the image/video lineup)
- [AI LoRA Trainer API](https://muapi.ai/lora-trainer) — train custom LoRA adapters on Flux, FLUX.2 Klein, Qwen-Image, and Z-Image from a zip dataset
- [Text to Speech API](https://muapi.ai/text-to-speech) — MiniMax Speech 2.6, Gemini TTS, and ElevenLabs single-voice and multi-speaker dialogue models
- [Gemini TTS API](https://muapi.ai/gemini-tts) — Google Gemini text-to-speech models
- [Lyria API](https://muapi.ai/lyria) — Google Lyria 3 Pro, Lyria 3, and Lyria 2 text-to-music models
- [Fish Audio API](https://muapi.ai/fish-audio-api) — Fish Audio S2.1 Pro long-form text-to-speech
- [ElevenLabs API](https://muapi.ai/elevenlabs-api) — ElevenLabs TTS Turbo 2.5 for natural single-voice narration and Text to Dialogue V3 for expressive multi-speaker audio
- [Speech to Text API](https://muapi.ai/speech-to-text) — Whisper transcription and Gemini Audio Vision audio analysis
- [Voice Cloning API](https://muapi.ai/voice-cloning) — MiniMax Voice Clone for text-to-speech and Suno Custom Voice Cloning for singing vocals
- [Video Upscaler API](https://muapi.ai/video-upscaler) — AI Video Upscaler, AI Video Upscaler Pro, and Topaz Video Upscale to 720p/1080p/2K/4K
- [Video to Audio API](https://muapi.ai/video-to-audio) — MMAudio video-to-audio and text-to-audio sound-effect and ambience generation
- [AI 3D Model API](https://muapi.ai/ai-3d-model-api) — Tripo3D and Meshy text-to-3D, image-to-3D, and multi-view-to-3D
- [Free AI APIs](https://muapi.ai/free-ai-apis) — watermarking, Gemini Omni voice/character creation, and Suno voice cloning — all $0.00 per call
- [Image Upscaler API](https://muapi.ai/image-upscaler) — general-purpose, SeedVR2 restoration, and Topaz premium super-resolution
- [Background Remover API](https://muapi.ai/background-remover) — image and video