kie-mcp
kie-mcp is an MCP server that gives Claude (or any MCP client) full access to kie.ai's 100+ generative AI models for images, video, and audio, with built-in cost control, model intelligence, and vertical playbooks.
Image generation (60+ models) — GPT Image 2.5, Nano Banana, FLUX 3, Seedream, Ideogram, Qwen, Grok Imagine; text-heavy layouts, characters, upscaling, background removal, transparent BG, inpainting.
Video generation (85+ models) — Veo 3.1, Kling, Seedance, Wan, Hailuo, HappyHorse, PixVerse, Runway; text-to-video, image-to-video, reference-to-video, lip-sync, avatars, 4K upscale, video extension.
Music & audio (20+ tools) — Suno generation, extend, cover, mashup, stems/vocal separation, MIDI export, lyrics, personas, plus ElevenLabs and Gemini TTS/dialogue/speech-to-text.
Vertical playbooks —
profile_briefreturns professional intake questions, model routing with live costs, prompt formulas, and workflows for 10 domains (architecture, game art, ads, product photography, film, brand, social video, audio branding, etc.).Smart model discovery —
list_modelsfilters across capabilities, descriptions, and embedded research (verdicts, weaknesses, competitor comparisons) to recommend the cheapest/best model.Cost control —
estimate_onlypreviews credits before submitting,max_creditscaps spending,spend_reporttracks actual charges,check_creditsshows balance.Async & long-running support —
wait: falsereturns a task_id;check_task/download_resultpoll and fetch results; files saved tokie/assets/raw/or a customdownload_dir.File handling & editing utilities —
upload_file(path, URL, or base64),grok_segment_map+grok_image_editfor region-targeted edits,seedream_layer_decomposeto split images into layers, Veo/Runway extend and upscale tools.Dual transport — stdio for local Claude Code/Desktop, or authenticated HTTP Streamable for remote clients.
Provides access to ByteDance's image and video generation models (e.g., Seedream, Seedance) through the kie.ai API.
Provides access to ElevenLabs' text-to-speech, audio isolation, and speech-to-text services through the kie.ai API.
Provides access to Google's image generation models (Nano Banana, Imagen) and video generation models (Veo) through the kie.ai API.
Provides access to Google Gemini's text-to-speech service with style-directed speech and multiple voices through the kie.ai API.
Provides access to Kuaishou's Kling video generation and AI avatar models through the kie.ai API.
Provides access to OpenAI's image generation models (GPT Image, GPT-4o Image) through the kie.ai API.
Provides access to Suno's music generation, sound effects, lyrics, and audio manipulation tools through the kie.ai API.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@kie-mcp Generate a 10-second video of a cat playing piano in Pixar style"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
kie-mcp
A comprehensive Model Context Protocol server for the kie.ai generation API. Gives Claude (and any MCP client) access to 60+ image models, 85+ video models, and 20+ audio tools with deep model intelligence built in.
Why this exists
Most MCPs are thin API wrappers. This one is different:
Vertical profiles (NEW in 5.0) — per-domain playbooks:
profile_briefreturns the intake questions a professional would ask, model routing per deliverable with live costs, per-model prompt formulas, and multi-tool workflows. Profiles double as MCP prompts (/kie-art:architecturein Claude Code). Verticals (10): architecture & interiors, video game assets, advertising & marketing, web & software product imagery, film & storyboarding, product photography & e-commerce, brand & graphic design, editorial & publishing, short-form social video, and audio branding & music.Deep research embedded — Every major model has a
researchfield with verdicts, prompt techniques, weaknesses, cost-efficiency analysis, and competitor comparisons. Researched by Averiguare, our model intelligence agent.Cost-aware — Every model has pricing in credits and USD. The MCP tells you the cheapest option for your use case.
Smart filtering —
list_models filter="lip sync"orfilter="architecture"orfilter="cheapest video"— searches across capability tags, descriptions, AND research fields.Dual-mode transport — stdio for local Claude Code, HTTP Streamable for remote Cowork/cloud usage.
Related MCP server: kie-ai-mcp-server
What you can do with it
Just ask Claude things like:
"Generate a brand presentation board for a perfume launch" — picks GPT Image 2 (best for text-heavy layouts)
"Make a 10s video of fruit scarecrows defending against crows, Pixar style" — recommends Veo 3.1 or Wan 2.7
"Generate music for a fantasy adventure game" — Suno V5
"Lip-sync this audio to my character image" — Kling AI Avatar or Infinitalk
"Upscale this video to 4K" — Veo 4K upscale or Topaz
"Replace the wall color in this room photo" — Flux Kontext Pro (best for surgical edits)
Model coverage
Image (60+)
OpenAI: GPT Image 2.6 Flare / Sunburst / Economy (NEW Nov 2026; Economy 5 cr flat, animated GIF), GPT Image 2.5 Flare / Sunburst (NEW — 1K-4K, transparent background), GPT Image 2, GPT-4o Image, GPT Image 1.5
Google: Nano Banana 2.1 (NEW, 4 cr), Nano Banana 2 / 2 Lite / Pro / Edit / Original, Imagen 4 (Fast/Standard/Ultra)
Black Forest Labs: FLUX 3 Klein T2I + Edit (NEW, 1.6 cr), Flux Kontext Pro/Max, Flux 2 Pro/Flex
ByteDance: Seedream 3.0 / 4.0 / 4.5 / 5.0 Lite
Alibaba: Wan 2.7 Image / Image Pro
Ideogram: v3, Character, Edit, Remix, Reframe
xAI Grok Imagine Image 2.0 (#2 Arena T2I + edit; free segment map → region-targeted edit chain; whole-image edits of ANY uploaded image)
ByteDance Seedream 5.0 Pro (NEW — T2I/I2I + layer decomposition: split any image into layer files)
Qwen Image 3.0 / 3.0 Pro (NEW — seed, negative prompts, 2K at the 1K price on standard)
Qwen Image 2.1 (NEW — transparent background, mask inpainting, 10-ref compositing)
Others: Qwen/Qwen2, Z-Image, Grok Imagine 1.x, Recraft, Topaz
Video (85+)
Google Veo 3.1: Quality / Fast / Lite (T2V + I2V), Extend, 1080p/4K upscale
Alibaba HappyHorse: 1.1 (NEW — T2V/I2V/R2V with native audio + 7-language lip-sync), 1.0 (T2V/I2V/R2V/Video Edit)
ByteDance Seedance: 2.5 (30s single takes, 1080p, draft→1080p upgrade) / 2.0 / 2.0 Fast / 2.0 Mini (1.5 Pro retired upstream Oct 2026)
Kuaishou Kling: 4.0 Flash (NEW — 720p, 17 cr/s), 4.0 (NEW — reference-to-video with named @subjects/voices/video refs, keyframe-to-video, 3-30s), 3.0 Omni "O3" (NEW — per-shot multi_prompt scripting, 4K, video Transformation), 3.0, 3.0 Turbo, 2.6, V2.5 Turbo, V2.1 Master/Pro/Standard, AI Avatar
Alibaba Wan: 3.0 + 3.0 Prime (NEW — unified prompt-or-media, audio), 2.7 (T2V/I2V/Edit/R2V), 2.6, 2.5, 2.2 Turbo, Animate
Google Gemini Omni: Video + 1.1 Flash (NEW — first→last-frame mode, 360p-4K; image/voice/video/character references)
MiniMax Hailuo: H3 (NEW — 2K + native stereo audio, image+video+audio references, first→last-frame I2V), 2.3 Pro/Standard, 02 Pro/Standard
xAI Grok Imagine: Video 1.5 preview (NEW — I2V with native audio, cheapest audio video), T2V, I2V, Upscale, Extend
Avatar / lip-sync: OmniHuman 2.0 (NEW — 14 cr/s, up to 5-minute audio), OmniHuman 1.5 (NEW — audio-driven full-body avatar + free subject-detection + human-identification utilities), Volcengine Video Lip-Sync (NEW — re-dub existing footage), Kling AI Avatar, Infinitalk
PixVerse V6 / V6 Max (V6 Max NEW: higher-quality T2V/I2V): T2V, I2V (viral templates), Transition (first→last morph), Fusion R2V (@ref_name), Extend — budget all-rounder with native audio
Runway: Aleph, Aleph Edit, Extend
Others: ByteDance V1 Pro/Lite, Topaz upscale
Audio (20+)
Suno: Music Gen, Extend, Cover, Add Instrumental/Vocals, Replace Section, Lyrics, Sounds, Sound Effects, MIDI, Music Video, Cover Art, Mashup, Persona, Timestamped Lyrics, Boost Style, Vocal Separation, WAV, Custom Voice cloning (experimental)
NVIDIA Nemotron 3.5 Transcribe (NEW):
speech_to_textdefault — 0.3 cr per minute, ≤60 min, 32 languages, word timings + speaker labels, translationElevenLabs: TTS (Turbo 2.5 + Multilingual V2), Text-to-Dialogue V3, Audio Isolation, Speech-to-Text
Google Gemini TTS (NEW): style-directed speech, 30 voices, 2-speaker dialogue, inline tone tags — ~4.2 cr/min
Utility
File upload (URL or base64)
Veo Extend, 1080p Upscale, 4K Upscale
Runway Extend
Task status, credit check, raw asset listing
Installation
Prerequisites
Node.js 18+
A kie.ai API key from kie.ai/api-key
Setup
git clone https://github.com/YOUR_USERNAME/kie-mcp.git
cd kie-mcp
npm installRun as stdio MCP (Claude Code, Claude Desktop)
Add to your Claude config (~/.claude.json for Claude Code, or your MCP client's equivalent):
{
"mcpServers": {
"kie-art": {
"command": "node",
"args": ["/absolute/path/to/kie-mcp/server.mjs"],
"env": {
"KIE_API_KEY": "your-kie-ai-api-key",
"KIE_PROJECT_ROOT": "/optional/path/for/outputs"
}
}
}
}Or use the Claude Code CLI:
claude mcp add -s user kie-art /usr/bin/env -- KIE_API_KEY=your-key node /path/to/server.mjsRun as HTTP MCP (Cowork, remote clients)
HTTP mode requires a bearer token and listens on 127.0.0.1 by default.
export KIE_MCP_AUTH_TOKEN=$(openssl rand -hex 32)
KIE_API_KEY=your-key node server.mjs --http --port=3100Configure your MCP client with the URL and the token:
{
"mcpServers": {
"kie-art": {
"type": "http",
"url": "http://127.0.0.1:3100/mcp",
"headers": { "Authorization": "Bearer <your KIE_MCP_AUTH_TOKEN>" }
}
}
}To reach it remotely, put it behind a tunnel (ngrok / Cloudflare Tunnel) or a reverse proxy. The server stays on loopback and the tunnel forwards to it. Add the tunnel's hostname to KIE_MCP_ALLOWED_HOSTS. Treat the token like a password: anyone who has it can spend your kie credits and upload media files from the server's disk.
Security note (5.2.2): earlier versions ran HTTP mode on
0.0.0.0with no authentication andAccess-Control-Allow-Origin: *. Any web page open in your browser, or anyone who had the tunnel URL, could call the tools, including reading local files throughupload_file. Upgrade if you use--http.
Environment variables
Variable | Required | Purpose |
| yes | Your kie.ai API key |
| no | Server-wide default for where generated files are saved (default: server cwd; files go to |
| no | Port for HTTP mode (default: 3100) |
| no |
|
| no |
|
| no | Default spending cap per tool call, in credits. Calls whose estimate exceeds it are refused before submission; a per-call |
| HTTP mode | Bearer token HTTP clients must send ( |
| no | HTTP bind address (default |
| no | Comma-separated extra |
| no | Comma-separated browser origins allowed to call the server (CORS). Requests with any other |
| no | Callback URL sent with Suno generation requests (kie.ai requires the field; results are fetched by polling regardless). Defaults to an inert placeholder — set this only if you want to receive the callbacks yourself |
| no | Max simultaneous task-creation calls (default 4). Excess parallel generations queue inside the server instead of hitting kie.ai's rate limits — parallel tool calls are safe |
| no | Blocking-mode polling budget per tool category, in seconds (defaults: 600 / 900 / 300 / 300). Per-call |
Tools available
generate_image, generate_video, generate_music, generate_sfx,
generate_tts, generate_gemini_tts, generate_dialogue, generate_sounds, generate_lyrics,
generate_persona, generate_mashup, generate_cover_art,
generate_midi, create_music_video,
prepare_voice_clone, create_voice_clone, regenerate_voice_clone,
create_omni_voice, create_omni_character,
extend_music, cover_audio, upload_extend_audio,
add_instrumental, add_vocals, replace_section,
convert_to_wav, separate_vocals, boost_style,
get_timestamped_lyrics, audio_isolation, speech_to_text,
profile_brief,
list_models, check_task, list_tasks, check_credits, spend_report,
download_result, list_raw_assets, upload_file,
grok_segment_map, grok_image_edit, seedream_layer_decompose,
veo_extend, veo_upscale_1080p, veo_upscale_4k, runway_extendCost control
kie bills per call, and a few calls are expensive (1080p/4K video, layer splits, 1080p draft upgrades). Every paid tool takes two optional parameters:
estimate_only: truereturns the estimated credits from published rates, including resolution/size tiers and video duration, without submitting anything.max_credits: Nrefuses to submit when the estimate exceedsN. SetKIE_MAX_CREDITS_PER_CALLto apply a default cap to every call.
generate_video { model: "bytedance/seedance-2-5", model_options: { resolution: "1080p", duration: 10 }, estimate_only: true }
→ 💰 ~1580 credits (~$7.90) — 158 cr/s × 10s. Nothing was submitted.spend_report shows where the credits went. It totals the server's task history over the last N days, by model or by day, using the actual amounts kie charged (it looks up any missing ones), and lists failed tasks separately.
Smart model recommendations
Try these queries in any MCP client:
list_models filter="reasoning" # GPT-4o, Nano Banana, GPT Image 2
list_models filter="lip-sync" # OmniHuman 1.5, Volcengine, Kling Avatar, HappyHorse 1.1
list_models filter="multi-shot" # Kling 3.0/Turbo
list_models filter="cheapest video" # Grok Imagine 1.5, Wan Flash
list_models filter="alibaba" # HappyHorse 1.0/1.1 family
list_models filter="best visual quality" # Veo Quality, Seedance 2.0
list_models filter="text rendering" # Ideogram v3, GPT Image 2
list_models filter="character" # Ideogram Character, Kling AI AvatarArchitecture
server.mjs # Transport, helpers, tool handlers (~2700 lines)
├── createMcpServer() # Factory for stdio + HTTP modes
├── Tool handlers # generate_*, list_*, etc.
└── helpers # polling, recovery, pricing, validation, download
data/ # Pure data, imported (and re-exported) by server.mjs
├── registry-image.mjs # MODEL_REGISTRY — image models (47+)
├── registry-video.mjs # VIDEO_MODEL_REGISTRY — video models (80+)
├── registry-audio.mjs # AUDIO_TOOLS_REGISTRY — audio tool metadata
├── pricing.mjs # PRICING, PRICING_ESTIMATED, PROMPT_CAPS
└── voices.mjs # ELEVENLABS_VOICES catalogThe registries and pricing live in data/*.mjs so model-catalog changes are reviewable diffs instead of edits buried in a 5000-line file; server.mjs imports and re-exports them (tests and downstream keep importing from server.mjs).
Each model entry has:
name,description,capabilities(tags),pricing(credits)aspectRatios,options(with types and defaults)buildBody/buildInput(request builders)research(Averiguare verdicts, prompt techniques, weaknesses, comparisons, sources)
Development
npm run check # node --check server.mjs (syntax)
npm test # offline unit tests for the pure helpers (test/*.test.mjs)
npm run smoke # live end-to-end over MCP stdio — needs KIE_API_KEY
# (spends ~0 credits; uses the free subject-detection model)server.mjs guards its side effects behind a main-module check, so it can be imported by tests (test/unit.test.mjs) to exercise the pure helpers without starting a server. test/harness.mjs is a reusable stdio JSON-RPC client for driving the real server in smoke/integration checks. CI (.github/workflows/ci.yml) runs the syntax check + unit tests on Node 20 and 22 for every push and PR.
Live check (run this weekly)
npm run live-check # KIE_API_KEY from env, else ~/.claude.json kie-artRuns the whole maintenance routine in about 2 minutes and writes live-check-report.md:
Docs-index diff: docs.kie.ai pages added or removed since the last run. This is the most reliable new-model signal.
Drift watch with live slug probes: price drift, paused/removed models, catalog changes.
Watch list: real calls to models that were failing upstream (e.g. ElevenLabs), so you know when to re-enable them. Failures cost 0 credits.
npm auditsummary.
Run it from your own machine. GitHub-hosted runners can't fetch kie.ai pages, so the scheduled workflow below only reports that it's blind.
Drift watch
kie.ai changes things without notice — advertised prices, model availability, even API shapes. .github/workflows/drift-watch.yml runs scripts/drift-watch.mjs weekly (and on demand) to scan for it: paused/removed slugs, pricing that no longer matches the PRICING table, and new models in kie's catalog. Findings land in a single rolling GitHub issue. Add a KIE_API_KEY repo secret to enable the per-slug liveness probes (0 credits — empty-input validation errors); the pricing and new-model scans need no secret. Run locally with node scripts/drift-watch.mjs.
Releasing
Releases are automated by .github/workflows/release.yml. To cut a release:
Bump the version in
package.json,server.json(both the top-levelversionandpackages[0].version), andserver.mjs(SERVER_INFO+ the/healthhandler), and add a## [X.Y.Z]section toCHANGELOG.md. Merge tomain.Tag and push:
git tag vX.Y.Z && git push origin vX.Y.Z
The workflow verifies the tag matches every in-repo version string, publishes to npm with provenance (NPM_TOKEN repo secret), and creates the GitHub Release using the matching CHANGELOG section as the notes. A tag whose version doesn't match the code fails fast without publishing. workflow_dispatch is an emergency manual publish of the current package.json version.
Credits
Built with the MCP TypeScript SDK
Powered by kie.ai — affordable unified API for 100+ AI models
Model intelligence by Averiguare — "No sabes hasta que averiguas — y averiguo en todas partes."
License
MIT
Available Tools
47 toolsadd_instrumentalC
Add instrumental backing to uploaded vocal audio via Suno.
Inputs: uploadUrl (required), tags, wait, model, +7 more.
add_vocalsC
Add AI vocals to uploaded instrumental audio via Suno.
Inputs: prompt (required), uploadUrl (required), wait, model, +8 more.
audio_isolationB
Isolate vocals or audio from background noise using ElevenLabs via kie.ai. Input an audio URL, get clean isolated audio back.
Inputs: audio_url (required), filename, max_credits, download_dir, +1 more.
boost_styleB
NEW — Convert concise style input (e.g. "Pop, Mysterious") into enhanced style description for music generation.
Inputs: content (required), max_credits, estimate_only.
check_creditsB
Check remaining kie.ai account credits
Inputs: no named parameters.
check_taskA
Check the status of a kie.ai generation task by taskId
Inputs: task_id (required).
convert_to_wavB
Convert a Suno track to lossless WAV format. Downloads to kie/assets/raw/.
Inputs: taskId (required), audioId (required), wait, filename, +4 more.
cover_audioC
Create an AI cover from uploaded audio — custom vocals, style, and instrumentation via Suno.
Inputs: uploadUrl (required), wait, model, style, +11 more.
create_music_videoB
Generate an MP4 music video visualization from a Suno track. Downloads to kie/assets/raw/.
Inputs: taskId (required), audioId (required), wait, author, +6 more.
create_omni_characterB
NEW — Create a reusable visual character for Gemini Omni video generation. Combines image + optional voice. Returns characterId.
Inputs: image_urls (required), descriptions (required), audio_ids, max_credits, +2 more.
create_omni_voiceA
NEW — Create a reusable voice character for Gemini Omni video generation. Returns kieAudioId for use in generate_video audio_ids.
Inputs: name (required), audio_id (required), max_credits, estimate_only, +2 more.
create_voice_cloneA
EXPERIMENTAL (#20) — STEP 2 — after the voice owner records the verification phrase from prepare_voice_clone, submit that recording to finish the voice. On success returns a voiceId usable in generate_music. Unverified end-to-end.
Inputs: task_id (required), verify_url (required), voice_name (required), wait, +4 more.
download_resultC
Download a completed task result to kie/assets/raw/
Inputs: task_id (required), filename, download_dir.
extend_musicB
Extend/continue an existing Suno track from a specific point. Requires audioId from a previous generate_music task.
Inputs: prompt (required), audioId (required), wait, model, +9 more.
generate_cover_artA
NEW — Generate album cover art image for an existing Suno music track. One call per taskId only.
Inputs: taskId (required), filename, download_dir.
generate_dialogueA
Generate multi-speaker dialogue using ElevenLabs Text-to-Dialogue V3 via kie.ai. Great for conversations between characters. Downloads to kie/assets/raw/.
Inputs: dialogue (required), wait, filename, stability, +5 more.
generate_gemini_ttsA
NEW — Google Gemini native TTS via kie.ai: style-directed speech from natural-language direction, 30 named voices, up to 2 speakers, inline tone tags like [whispers]/[laughs] (flash model). ~4.2 credits per MINUTE of audio — cheaper than all ElevenLabs tiers. Simple mode: pass text (+ optional voice_name). Dialogue mode: pass speakers + dialogue_turns. model=flash is most expressive (keep expected audio <60s — quality degrades on long takes); model=pro is more stable for multi-minute narration. Downloads to kie/assets/raw/.
Inputs: text, wait, model, scene, +10 more.
generate_imageA
Generate an image using kie.ai. TIP: for architecture/game-art/advertising/product-UI jobs, call profile_brief first — it returns the vertical's intake questions, routing, and prompt formulas. (60+ models). Downloads to kie/assets/raw/. MODEL GUIDE: Architecture/blueprints→gpt4o or nano-banana-2 (reasoning, 8cr). Cheap fast all-rounder→nano-banana-2-1 (4cr, ~13s, NEW Oct 2026; NB2 doubled to 8cr). Game art/3D→seedream/4.5 or 5-lite. Character sheets→ideogram/character. Text/logos→ideogram/v3 (best text). Photo editing→flux-kontext-pro. Newest OpenAI→gpt-image-2-5/flare-* (6cr @1K, fast default) or gpt-image-2-5/sunburst-* (premium polish); both 1K-4K + transparent background (NEW). Split any image into layers→seedream_layer_decompose tool (7cr/layer, NEW). Generate-then-refine by named region→grok-imagine-image-2-0/text-to-image (4cr, #2 Arena T2I+edit) then grok_segment_map (free) + grok_image_edit (4cr; also edits ANY uploaded image via image_urls mode). Anime→qwen (3cr cheapest); qwen2-1/* (4cr, NEW) adds transparent BG, mask inpainting, 10-ref compositing. Fast drafts→nano-banana-2-lite (4cr, ~4s, NEW). Upscale→recraft/crisp-upscale (0.5cr). BG removal→recraft/remove-background. Cheapest→flux-3/klein-text-to-image (1.6cr @1K, 4K up to 6.5cr, NEW) or flux-3/klein-edit-image (1.6cr, 10 free refs); z-image,qwen (3cr). Best quality→nano-banana-pro (24cr), flux-kontext-max (100cr). Use list_models filter="use-case" to explore.
Inputs: prompt (required), wait, model, filename, +7 more.
generate_lyricsA
Generate song lyrics from a prompt using Suno AI (max 200 characters). Returns text, no file download.
Inputs: prompt (required), max_credits, estimate_only.
generate_mashupC
NEW — Mashup up to 2 Suno tracks into one new track. Provide audioIds from previous generations.
Inputs: audioIds (required), wait, model, prompt, +6 more.
generate_midiA
Export a Suno track to MIDI notation. Downloads .mid file to kie/assets/raw/.
Inputs: taskId (required), wait, audioId, filename, +4 more.
generate_musicA
Generate music using Suno via kie.ai. Supports V5.5 (custom style), V5 (best quality), V4.5+, V4.5, V4. Up to 8 minutes. Great for game music stems, ambient tracks, and jingles. Polls until done and downloads to kie/assets/raw/.
Inputs: prompt (required), wait, model, style, +7 more.
generate_personaA
NEW — Create a Suno Persona (reusable music character) from an existing Suno track. Requires taskId from V3.6+ generation.
Inputs: name (required), taskId (required), audioId (required), description (required), +3 more.
generate_sfxA
Generate a sound effect from text via Suno V5 (kie.ai removed the ElevenLabs sound-effect model). Great for game sounds: UI clicks, magic spells, item pickups, explosions. For loop/BPM/key control use generate_sounds instead. Downloads to kie/assets/raw/.
Inputs: text (required), wait, filename, max_credits, +5 more.
generate_soundsC
Generate loopable sound effects with BPM, key, and loop control via Suno. Downloads to kie/assets/raw/.
Inputs: prompt (required), wait, model, filename, +8 more.
generate_ttsB
Generate speech from text using ElevenLabs via kie.ai. Supports Turbo 2.5 (fast) and Multilingual V2 (high quality). Downloads to kie/assets/raw/.
Inputs: text (required), wait, model, speed, +7 more.
generate_videoA
Generate a video using kie.ai (85+ models). Downloads to kie/assets/raw/. MODEL GUIDE: Best cinematic→veo-3/text-to-video (50cr/s, audio). Fast+cheap→grok-imagine-video-1-5-preview (1.6-3cr/s, audio, NEW), wan/flash-image-to-video (6-8cr/s measured; alias of wan/2-6-flash). Budget cinematic→hailuo-standard (4cr/s). First→last-frame or anything-from-anything refs→gemini-omni/flash-1-1 (NEW, est. ~63cr per 4s clip). Budget multimodal refs→bytedance/seedance-2-mini (9.5cr/s @480p). 30s single takes→bytedance/seedance-2-5 (now up to 1080p; set model_options.draft=true for a 28cr/s 480p preview, then upgrade the approved take with bytedance/seedance-2-5-draft). Cheap or 4K image-to-video with audio→vidu-q4/image-to-video (7.2cr/s @540p … 62.4 @4K, NEW); 15-image refs→vidu-q4/reference-to-video. Budget all-rounder w/ audio+templates+extend→pixverse-v6 family (4-9.6cr/s, NEW; I2V is its strength; transition=first/last-frame morph). Higher-quality PixVerse without templates→pixverse-v6-max T2V/I2V (NEW, 8.8cr/s @720p). Multilingual lip-synced dialogue→happyhorse-1-1 T2V/I2V/R2V (NEW). 2K + stereo audio→minimax-h3 (8cr/s @768P, price halved Sept 2026). Per-shot scripted multi-shot→kling-3-omni (14cr/s @720p; transformation=restyle existing video). Long consistent-character clips (3-30s) with named @subjects, voices and reference videos→kling-4/reference-to-video (NEW, 28.5cr/s @720p); first→keyframes→last animation→kling-4/keyframe-to-video (NEW). Next-gen Wan draft→wan/3-0-video (8cr/s @480P, NEW). Fast Kling→kling/v3-turbo (18cr/s, audio, NEW). Image-to-video→veo-3/image-to-video (include a sound cue like "SFX: room tone" — Veo I2V intermittently fails its audio pass without one; images must be served with their real Content-Type, upload via upload_file), kling/image-to-video. Avatar/talking head→omnihuman-2-0 (NEW Oct 2026: 14cr/s, up to 5-min audio, half of 1.5's price), omnihuman-1-5 (multi-person masks, 1080p), kling/ai-avatar-pro, infinitalk/from-audio. Re-dub existing footage→volcengine/video-to-video-lip-sync (8cr/s, NEW). Motion control→kling/motion-control, wan/animate-move. Extend video→use veo_extend or runway_extend tools. NOTE: Sora 2 family removed (OpenAI API sunset Sept 2026). Use list_models filter="use-case" to explore.
Inputs: prompt (required), wait, model, filename, +7 more.
get_timestamped_lyricsB
NEW — Get word-level timestamped lyrics from a Suno track. Useful for karaoke, captioning, or sync.
Inputs: taskId (required), audioId (required), max_credits, estimate_only.
grok_image_editA
Edit an image with Grok Imagine Image 2.0 (4 credits). TWO MODES: (1) Region mode — pass task_id (a prior Grok 2.0 generation) + mask_indexs from grok_segment_map (run it first, free, pick regions by NAME): only those regions change. (2) Whole-image mode (NEW Aug 2026) — pass image_urls (ANY uploaded/external image, e.g. from upload_file) + aspect_ratio + prompt: instruction-based edit of the full image, no segmentation. Returns a new full image; result task_ids chain back into segment/edit for iterative refinement. Downloads to kie/assets/raw/.
Inputs: prompt (required), task_id, filename, image_urls, +5 more.
grok_segment_mapA
FREE (0 credits). Segment a Grok Imagine Image 2.0 generation into NAMED regions for targeted editing. Returns each region's index, semantic name (e.g. "red apple", "wooden table"), and mask PNG URL. Workflow: generate_image model="grok-imagine-image-2-0/text-to-image" → grok_segment_map (this, free) → grok_image_edit with the mask_indexs you want changed. Only works on task_ids from a Grok Image 2.0 generation.
Inputs: task_id (required).
list_modelsA
List all available kie.ai models with their aspect ratios and model-specific options
Inputs: filter, verbose.
list_raw_assetsA
List all files in kie/assets/raw/ waiting to be processed
Inputs: no named parameters.
list_tasksB
List recent image generation tasks from this session
Inputs: limit.
prepare_voice_cloneA
EXPERIMENTAL (#20) — STEP 1 of Suno custom-voice cloning (FREE). Submit a clean vocal sample; polls to wait_validating, after which kie sends a verification phrase to your KIE_CALLBACK_URL for the voice owner to read aloud (then use create_voice_clone). ⚠️ Only clone a voice you OWN or have explicit permission to use. The completion step is not verified end-to-end and needs a real callback URL.
Inputs: voice_url (required), wait, language, vocal_end_s, +2 more.
profile_briefA
Get a vertical playbook before generating: the intake questions a professional in that domain would ask, model routing per deliverable with live costs, per-model prompt formulas, and multi-tool workflows. Call with no args to list available profiles. Covers image, video, and audio verticals (architecture, game assets, advertising, product photography, film, brand, web product, editorial, short-form social video, audio branding). Call this FIRST when the user's request belongs to a known vertical — then ask the user only the unanswered intake questions, conversationally.
Inputs: profile, request.
regenerate_voice_cloneB
EXPERIMENTAL (#20) — retry a failed/incomplete custom-voice task by its task_id.
Inputs: task_id (required), max_credits, estimate_only.
replace_sectionC
Replace a time range in a Suno track with new AI-generated content.
Inputs: prompt (required), taskId (required), audioId (required), infillEndS (required), +11 more.
runway_extendC
Extend an existing Runway Aleph video with continuation content.
Inputs: prompt (required), task_id (required), quality, filename, +3 more.
seedream_layer_decomposeA
Split ANY image into independent layers with Seedream 5.0 (billed per OUTPUT layer incl. base). tier="pro" (default): 7 cr @1K, 14 @2K, so a 3-layer split ≈ 21 cr. tier="flash" (NEW): 3.24 cr/layer at any size, so ≈ 10 cr. Works on any public image URL (upload local files with upload_file first). Describe which elements become layers in the prompt, optionally bounding them with x1 y1 x2 y2 tags. Downloads every layer image to kie/assets/raw/.
Inputs: image_url (required), size, tier, prompt, +5 more.
separate_vocalsC
Separate vocals from instrumentals, or split into individual stems. Downloads to kie/assets/raw/.
Inputs: taskId (required), audioId (required), type, wait, +5 more.
speech_to_textB
Transcribe audio to text using ElevenLabs Scribe v1. Supports diarization and audio event tagging. Returns transcription text.
Inputs: audio_url (required), diarize, max_credits, estimate_only, +2 more.
spend_reportA
Where did the credits go? Totals kie.ai spend from this server's task history (kie/assets/task-history.jsonl under the project root) over the last N days, grouped by model or day. Uses the ACTUAL credits kie charged when known, looks up missing ones from kie (refresh), and falls back to published prices (marked "est."). Failed tasks are listed separately (kie normally refunds them).
Inputs: days, refresh, group_by.
upload_extend_audioA
NEW — Extend uploaded audio (NOT a Suno track) with new AI-generated content. For Suno tracks, use extend_music instead.
Inputs: uploadUrl (required), continueAt (required), wait, model, +9 more.
upload_fileA
Upload a file to kie.ai and get a public URL back. Image files are named by their REAL format (magic bytes), so a JPEG saved as .png is uploaded as .jpg — kie serves files with a Content-Type taken from the name, and Veo I2V rejects a mismatch. Use this to upload local images/audio/video before passing them to generation tools (image-to-image, image-to-video, reference/ingredient inputs). PREFER file_path for local files. Files expire after 3 days (kie temp storage).
Inputs: file_url, file_name, file_path, base64_data, +1 more.
veo_extendB
Extend an existing Veo 3.1 video with additional content. Requires taskId from a previous Veo generation.
Inputs: prompt (required), task_id (required), model, seeds, +4 more.
veo_upscale_1080pB
Upscale a Veo 3.1 video to 1080p resolution. Requires taskId from a completed Veo generation.
Inputs: task_id (required), index, filename, max_credits, +2 more.
veo_upscale_4kA
Upscale a Veo 3.1 video to 4K resolution. Takes 5-10 minutes. Requires taskId from completed Veo generation.
Inputs: task_id (required), index, filename, max_credits, +2 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 47 tools
Core generation tools (generate_image, generate_video, generate_music, generate_tts) are clearly separated by modality, but the audio cluster has substantial overlap: generate_sfx vs generate_sounds, generate_tts vs generate_gemini_tts vs generate_dialogue, and six Suno editing tools (extend_music, upload_extend_audio, cover_audio, add_vocals, add_instrumental, replace_section) that an agent could easily confuse. The descriptions do disambiguate most of these cases (e.g. 'for loop/BPM/key control use generate_sounds instead'), which keeps it above a 2, but selection still requires reading long guidelines.
The dominant pattern is verb_noun (generate_image, list_tasks, check_credits, extend_music, upload_file), with model-specific tools consistently prefixed (veo_*, grok_*, runway_*). A handful of deviations exist — audio_isolation, speech_to_text, seedream_layer_decompose, profile_brief use noun-first or noun_verb ordering — but overall the vocabulary is predictable and readable.
47 tools is heavy for a single server, and many of the additions are narrow Suno/experimental features (generate_mashup, boost_style, regenerate_voice_clone, get_timestamped_lyrics, generate_midi, create_music_video) that inflate the surface well beyond core generate/list/check workflows. The breadth of modalities justifies more than a dozen tools, but this exceeds what an agent can reliably navigate.
Coverage is unusually thorough across image, video, music, TTS, SFX, voice cloning, transcription, upscaling, extending, segmentation, uploads, credit checks, and spend reporting, with clear lifecycle chains (generate → check_task → download_result). Minor gaps remain: no task cancellation, no asset deletion/cleanup, and task listing appears scoped to the current session only.
Maintenance
Related MCP Connectors
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Chat, image, video, and audio models from many vendors behind one remote MCP server and one key.
Official hosted MCP server for Wiro AI. Run 1,000+ image, video, audio and LLM models, build and run workflows that chain them, and deploy and talk to AI agents from any MCP client.
Generate images, video, audio and short films with 140+ AI models from any MCP client.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA Model Context Protocol server that enables AI assistants to perform comprehensive video and audio editing operations including trimming, effects, overlays, audio processing, and YouTube downloads.52 PyPI30MIT
- AlicenseNot gradedqualityDmaintenanceUniversal MCP server for all Kie AI image generation and editing models, enabling text-to-image, image-to-image, and image editing via natural language.2MIT
- AlicenseCqualityDmaintenanceA FastMCP-based Model Context Protocol (MCP) server that provides unified access to multiple AI APIs including OpenAI GPT, Google Gemini, Anthropic Claude, and xAI Grok.58 npmMIT
- FlicenseNot gradedqualityDmaintenanceSelf-hosted MCP server that integrates Kie.ai image and video generation into GoHighLevel's AI tools, enabling AI agents to create media on demand.1-