# APImage > APImage is an AI image and video generator for creating UGC videos and ads, viral videos for brands and social media, and content for faceless channels. APImage is a web app at https://apimage.org, with a REST API and an MCP server. It is built and operated in Germany. Sign up at https://app.apimage.org/signup. ## What APImage is for The four jobs people hire it for, in order of how often: 1. **UGC-style ad video.** A product photo plus a line of direction becomes a creator-style ad, without a creator, a phone or a shoot. 2. **Viral short-form video for brands and social.** Vertical clips for TikTok, Reels and Shorts, generated in bulk so hooks can be tested rather than guessed. 3. **Faceless channel content.** Video and images for channels that publish daily and never show a presenter. 4. **Product and marketing imagery.** Product shots, backgrounds, thumbnails, posters and on-model photography from a single reference image. ## Video generation The studio carries 18 video models. The REST API and the MCP server carry two families of them, FLUX 3 Video and Seedance, across five modes: text-to-video, image-to-video, video-to-video, lip sync and draft-enhance. API generation is asynchronous: a call returns an id and the exact credit cost, then you poll or take a webhook. Clip length and resolution are bounded per model, so they are stated per model rather than as one range: - **FLUX 3 Video** (Black Forest Labs), from 114 credits per 5s, starter and up, also in the API and MCP. Video with synchronized sound, up to Full HD and 20 seconds. The one you ship ads with. - **FLUX 3 Video Draft** (Black Forest Labs), from 34 credits per 5s, starter and up, also in the API and MCP. The same take at a third of the price. Explore fast, then re-render the keeper in Full HD. - **Seedance 2.5** (ByteDance), from 184 credits per 5s, starter and up, also in the API and MCP. The long take. Up to 30 seconds in one generation, with native audio and 30 reference images. - **Seedance 2.0** (ByteDance), from 115 credits per 5s, starter and up, also in the API and MCP. Cinematic motion with native audio, up to 4K. Steer it with nine references or start and end frames. - **Seedance 2.0 Fast** (ByteDance), from 91 credits per 5s, starter and up, also in the API and MCP. Seedance 2.0 at lower latency and price, up to 720p. Built for testing hooks in bulk. - **Seedance 2.0 Mini** (ByteDance), from 61 credits per 5s, starter and up, also in the API and MCP. The cheapest Seedance. Quick drafts with sound at 480p or 720p. - **Veo 3.1** (Google), from 267 credits per 8s, starter and up. Google's flagship. Lifelike motion with dialogue and sound, up to 4K. - **Veo 3.1 Fast** (Google), from 134 credits per 8s, starter and up. Veo quality and sound at a fraction of Veo 3.1's price. - **Veo 3.1 Lite** (Google), from 40 credits per 8s, starter and up. A cheap way to draft with Veo's motion and sound. Iterate here, finish on Veo 3.1. - **Kling 4.0** (Kuaishou), from 131 credits per 5s, starter and up. The newest Kling. Up to 30 seconds with stereo sound in every clip, and up to ten reference photos to hold a cast or product. - **Kling 3.0** (Kuaishou), from 117 credits per 5s, starter and up. Cinematic motion with native audio, up to 15 seconds, with start and end frames. - **Kling O3 Reference** (Kuaishou), from 117 credits per 5s, starter and up. Put up to four reference people or products into one shot and call them from the prompt. - **Kling 2.6 Pro** (Kuaishou), from 59 credits per 5s, starter and up. The reliable all-rounder: sound, start and end frames, five or ten seconds. - **Kling 2.5 Turbo** (Kuaishou), from 59 credits per 5s, starter and up. Fast, affordable motion with start and end frames. Silent. - **Wan 2.7** (Alibaba), from 84 credits per 5s, starter and up. Open-model motion with an automatic soundtrack, plus start and end frames. - **Wan 3.0** (Alibaba), from 42 credits per 5s, starter and up. Top of the video-edit leaderboards: images, clips and sound as references for new shots, edits and recasts. - **Hailuo 2.3** (MiniMax), from 47 credits per 6s, starter and up. Expressive characters and convincing physics, at 768p or 1080p. - **PixVerse V6** (PixVerse), from 21 credits per 5s, starter and up. Snappy social clips, stylised looks and start-to-end transitions. The cheapest way to test a hook. - FLUX 3 Video renders 5 to 20 seconds, at `hd` (720p) or `fhd` (1080p), and generates a synchronized audio track with the clip. - Seedance 2.0 renders 4 to 15 seconds at 480p to 4K, Seedance 2.0 Fast and Mini up to 720p, and Seedance 2.5 up to 30 seconds at up to 1080p. All four generate native audio. - Every price above is the model's cheapest setting. Length, resolution, sound and quality raise it, and the exact cost is shown before anything is generated. - Aspect ratios on Seedance: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 and `auto`. FLUX 3 Video adds 2:1. ## Image generation and editing - **FLUX 3 Image** (Black Forest Labs), 4 to 8 credits, every plan, also in the API and MCP. The newest FLUX. Precise edits, up to 10 references in one frame, and stills that match FLUX 3 Video. - **GPT Image 2.5 Flare** (OpenAI), 1 to 204 credits, every plan, also in the API and MCP. The best text rendering there is, now at half the wait. Thumbnails, posters, packaging. - **GPT Image 2.5 Sunburst** (OpenAI), 1 to 204 credits, every plan, also in the API and MCP. Holds your subject and your edits across turn after turn. Campaign creative. - **GPT Image 2** (OpenAI), 1 to 69 credits, every plan, also in the API and MCP. The previous OpenAI generation. Still strong on text and packaging. - **Flux 2 Max** (Black Forest Labs), 6 to 13 credits, plus and up, also in the API and MCP. Maximum detail and prompt adherence. Your hero shots live here. - **Flux 2 Pro** (Black Forest Labs), 3 to 6 credits, hobby and up, also in the API and MCP. High quality at a working price. The everyday workhorse. - **Flux 2 Klein 9B** (Black Forest Labs), 2 credits, every plan, also in the API and MCP. Fast and reliable. Ideal for iterating before you commit. - **Nano Banana Pro** (Google), 25 to 50 credits, hobby and up. Google's top image model. Photoreal, legible text and precise multi-image edits, up to 4K. - **Nano Banana 2** (Google), 10 to 27 credits, hobby and up. The fast Gemini image model, with edits, extreme formats down to 8:1 and up to 4K. - **Nano Banana** (Google), 7 credits, hobby and up. Quick, cheap edits and generations on Gemini 2.5 Flash Image. - **Seedream 5 Lite** (ByteDance), 6 credits, hobby and up. The newest Seedream. Dense layouts and text, and 4K for 6 credits. - **Seedream 4.5** (ByteDance), 7 credits, hobby and up. Crisp 2K and 4K images and multi-reference edits for a few credits. - **Ideogram 3** (Ideogram), 5 to 15 credits, hobby and up. Posters, logos and anything with words in it, with more than sixty looks to start from. - **Recraft V4.1** (Recraft), 6 to 35 credits, hobby and up. Illustration, brand graphics and clean design compositions. Images generate up to 4K, with transparent PNG output and reference images to hold a product or a face consistent across shots. Editing covers inpainting, object removal, background removal and replacement, frame extension and upscaling. ## Lip sync, voice, music and video tools - Lip sync: Kling Avatar 2 turns a portrait plus a recording, or a typed script read by an ElevenLabs voice or a cloned voice, into up to 60 seconds of talking video. Kling Lip Sync and VEED Lip Sync put new audio on a clip you already have. - Text-to-speech runs on ElevenLabs v3, Multilingual v2 and Turbo v2.5. Chatterbox speaks in a voice cloned from a short sample you upload. ### Photo tools - **Photo Relight** (Google), 14 credits, hobby and up. Move the light in a finished photo: direction, height, hardness, brightness and colour. Nothing else changes. - **Photo Angles** (Alibaba), 6 credits, hobby and up. The same shot from another camera position. Orbit the subject, raise or lower the camera, zoom in or out. ### Lip sync and avatars - **Kling Avatar 2** (Kuaishou), 94 credits per 10s, starter and up. A portrait plus a voice becomes a talking video with matching lips. Bring a recording, or type a script. - **Kling Avatar 2 Pro** (Kuaishou), 192 credits per 10s, starter and up. The talking avatar at higher fidelity, for faces and expressions that hold up at full size. - **Kling Lip Sync** (Kuaishou), 12 credits per 5s, starter and up. Make the person in a short clip say new audio. - **VEED Lip Sync** (VEED), 117 credits per 10s, starter and up. Lip sync for longer clips, up to a minute. ### Voice, music and sound - **ElevenLabs v3** (ElevenLabs), 17 credits per 1,000 characters, hobby and up. The most expressive voices. Direct the read with tags like [whispers] and [laughs]. - **ElevenLabs Multilingual v2** (ElevenLabs), 17 credits per 1,000 characters, hobby and up. Stable, natural narration in 29 languages, with fine control over the voice. - **ElevenLabs Turbo v2.5** (ElevenLabs), 9 credits per 1,000 characters, hobby and up. Half the price and fast. Good for drafts and long scripts. - **Chatterbox** (Resemble AI), 5 credits per 1,000 characters, hobby and up. Speak in your own voice, cloned from a short sample you upload. - **ElevenLabs Voice Changer** (ElevenLabs), 50 credits per minute, hobby and up. Re-voice a recording and keep its timing, emotion and delivery. - **ElevenLabs Dubbing** (ElevenLabs), 100 credits per minute, starter and up. Translate the speech in a video into another language, in the same voices. - **MiniMax Music 2.6** (MiniMax), 25 credits per song, hobby and up. Full songs with vocals. Describe the style, bring your lyrics or let it write them. - **Eleven Music** (ElevenLabs), 100 credits per minute, hobby and up. Studio-grade tracks at the exact length you need, up to five minutes. - **Lyria 3** (Google), 7 credits per clip, hobby and up. Thirty-second clips with vocals. Quick and cheap for trying ideas. - **Eleven Sound Effects** (ElevenLabs), 1 to 8 credits, hobby and up. Any sound from a description: footsteps, whooshes, ambience, UI clicks. - **Kling Video to Audio** (Kuaishou), 6 credits per clip, starter and up. Adds matching sound effects and music to a silent clip. - **VEED Clean Audio** (VEED), 3 credits per minute, hobby and up. Studio-quality speech from a noisy recording or video. ### Motion and swaps - **Kling 3.0 Motion Control** (Kuaishou), from 210 credits per 10s, starter and up. Motion Control: your character does every move of a video. Also behind Face Lock and camera copies. - **Kling 2.6 Motion Control** (Kuaishou), from 117 credits per 10s, starter and up. The cheaper motion transfer: copy a clip's moves, or its camera move, onto your photo. - **DreamActor v2** (ByteDance), 84 credits per 10s, starter and up. The lowest-priced motion transfer. Best with cartoon, non-human or several characters. - **Kling O3 Edit** (Kuaishou), from 234 credits per 10s, starter and up. Change what is in a clip: recast the lead, swap the outfit or restyle the look, and keep every move. - **PixVerse Swap** (PixVerse), from 50 credits per 10s, starter and up. Replace the person, an object or the background in a clip. - **MiniMax H3 Max** (MiniMax), from 500 credits per 10s, starter and up. Recast: swap the person in your video for yours. The face holds through turns and cuts; scene, moves and sound stay. - **Wan 2.7 Edit** (Alibaba), from 167 credits per 10s, starter and up. Edit or recast a clip with a prompt and one photo, keeping its sound. ### Video tools - **PixVerse V6 Extend** (PixVerse), from 21 credits per 5s added, starter and up. Continue a clip for up to 15 more seconds. - **Veo 3.1 Fast Extend** (Google), 117 credits per 7s, starter and up. Seven more seconds in Veo quality, with sound if you want it. - **LTX 2.3 Reframe** (Lightricks), from 167 credits per 10s, starter and up. Change a clip's frame shape and fill the new areas to match. - **ByteDance Video Upscaler** (ByteDance), from 12 credits per 10s, starter and up. Sharpen and upscale a clip to 1080p, 2K or 4K, optionally at 60 fps. - **Bria Video Background Removal** (Bria), 84 credits per 10s, starter and up. Cut the subject out of a clip, onto transparency or a solid colour. - **Depth Anything Video** (Depth Anything), 67 credits per 10s, starter and up. Turn a clip into a depth map, for compositing and 3D effects. ## Plans and pricing Credits are the single currency for everything. Prices are USD per month. | Plan | Monthly | Credits/month | Parallel video jobs | | --- | --- | --- | --- | | Free | $0 | 0 | 0 | | Hobby | $9 | 150 | 0 | | Starter | $19 | 300 | 2 | | Plus | $59 | 1,200 | 6 | | Ultra | $129 | 3,000 | 8 | | Team | $79 | 1,000 | 8 | | Scale | $245 | 2,500 | 8 | | Enterprise | Contact | Custom | 8 | - The free plan gives 3 watermarked images and no video. - Video generation starts on Starter, from $19 per month. - Paid plans carry full commercial rights, including client work and paid advertising. - Up to 8 videos and 8 images generate in parallel on the top tiers. - Pay-as-you-go overage is available once a monthly pool is spent. ## API, MCP and integrations - REST API, one key, nine endpoints, with webhooks and automatic credit refunds when a generation fails. - MCP server at https://mcp.apimage.org/mcp, exposing 20 tools over one Streamable HTTP endpoint, for Claude, Cursor, ChatGPT, Codex and any MCP client. - Native integrations for Zapier, Pipedream and n8n. - Video generation returns 403 on the free tier. The API caps three in-flight video jobs per account; a fourth returns 429 until one finishes. ## Core pages - [Home](https://apimage.org/): What APImage makes and who it is for - [AI video generation](https://apimage.org/video-generation): Text-to-video, image-to-video and video-to-video, with the model lineup - [AI UGC and ad video](https://apimage.org/ugc-ads): Creator-style ad video from a product photo - [Instagram comment-to-DM automation](https://apimage.org/instagram-dm-automation): Auto-DM people who comment a keyword on a Reel; free for 1 account and 1,000 messages a month - [AI lip sync](https://apimage.org/lip-sync): Make a photo talk, or put new words in a clip - [AI image generation](https://apimage.org/generate): Studio-grade images from a prompt, up to 4K - [AI image editing and inpainting](https://apimage.org/inpaint): Change one part of an image without touching the rest - [Remove background](https://apimage.org/remove-background): Transparent cutouts at a flat 2 credits per image - [AI models](https://apimage.org/models): All 63 video, image, lip sync, voice and music models, with credit costs - [AI model comparisons](https://apimage.org/compare): Head-to-head comparisons of models APImage runs, with specs, prices and which to pick when - [Pricing](https://apimage.org/pricing): Plans, credits, overage and what each tier unlocks ## Models, one page each - [Seedance 2.0](https://apimage.org/models/seedance-2-0): ByteDance. Run ByteDance Seedance 2.0 on APImage. Native audio, 480p to 4K, 4 to 15 seconds, text or image to video, 9 reference images plus reference clips and audio. From 115 credits per 5 seconds. - [Seedance 2.5](https://apimage.org/models/seedance-2-5): ByteDance. Run ByteDance Seedance 2.5 on APImage. Up to 30 seconds in one generation with native audio, 30 reference images, 10 clips and 10 audio tracks, at 480p to 1080p. From 184 credits per 5 seconds. - [FLUX 3 Video](https://apimage.org/models/flux-3-video): Black Forest Labs. Run FLUX 3 Video from Black Forest Labs on APImage. Up to Full HD, 5 to 20 seconds, with native synchronized audio and multilingual dialogue. From 114 credits per 5 seconds. - [FLUX 3 Video Draft](https://apimage.org/models/flux-3-video-draft): Black Forest Labs. Run FLUX 3 Video Draft on APImage. The FLUX 3 Video model at HD for a third of the price, 5 to 20 seconds with synchronized sound. From 34 credits per 5 seconds. - [Veo 3.1](https://apimage.org/models/veo-3-1): Google. Run Google Veo 3.1 on APImage. Lifelike motion with dialogue and native sound, 720p to 4K, 4 to 8 seconds, start and end frames. From 267 credits per 8 seconds. - [Veo 3.1 Fast](https://apimage.org/models/veo-3-1-fast): Google. Run Google Veo 3.1 Fast on APImage. Veo motion with dialogue and sound at 720p, 1080p or 4K, 4 to 8 seconds, start and end frames. From 134 credits per 8 seconds. - [Veo 3.1 Lite](https://apimage.org/models/veo-3-1-lite): Google. Run Google Veo 3.1 Lite on APImage. Veo motion with sound at 720p or 1080p, 4 to 8 seconds, start and end frames. From 40 credits per 8 seconds. - [Kling 4.0](https://apimage.org/models/kling-4-0): Kuaishou. Run Kling 4.0 on APImage. Kuaishou's newest video model: 3 to 30 seconds at 720p or 1080p with stereo sound and dialogue, up to 10 reference images, start and end frames, 21:9. From 131 credits per 5 seconds. - [Kling 3.0](https://apimage.org/models/kling-3-0): Kuaishou. Run Kling 3.0 on APImage. Cinematic motion with native audio, 3 to 15 seconds at 720p or 1080p, start and end frames, multi-shot. From 117 credits per 5 seconds. - [Kling O3 Reference](https://apimage.org/models/kling-o3-reference): Kuaishou. Run Kling O3 Reference on APImage. Put up to 4 reference people or products into one shot, 3 to 15 seconds at 720p or 1080p, with native audio. From 117 credits per 5 seconds. - [Kling 2.6 Pro](https://apimage.org/models/kling-2-6-pro): Kuaishou. Run Kling 2.6 Pro on APImage. Reliable motion with optional audio, 5 or 10 seconds, start and end frames, 16:9, 9:16 or 1:1. From 59 credits per 5 seconds. - [Kling 2.5 Turbo](https://apimage.org/models/kling-2-5-turbo): Kuaishou. Run Kling 2.5 Turbo on APImage. Fast, affordable silent video, 5 or 10 seconds, start and end frames, 16:9, 9:16 or 1:1. From 59 credits per 5 seconds. - [Wan 2.7](https://apimage.org/models/wan-2-7): Alibaba. Run Alibaba Wan 2.7 on APImage. Video with an automatic soundtrack, 2 to 15 seconds at 720p or 1080p, start and end frames, five aspect ratios. From 84 credits per 5 seconds. - [Wan 3.0](https://apimage.org/models/wan-3-0): Alibaba. Run Alibaba Wan 3.0 on APImage. Video from up to 9 reference images, 5 clips and 5 audio tracks, 2 to 30 seconds at 480p to 1080p, with sound. From 42 credits per 5 seconds. - [Hailuo 2.3](https://apimage.org/models/hailuo-2-3): MiniMax. Run MiniMax Hailuo 2.3 on APImage. Expressive characters and convincing physics, 6 or 10 seconds at 768p or 6 seconds at 1080p, from text or a start frame. From 47 credits per 6 seconds. - [PixVerse V6](https://apimage.org/models/pixverse-v6): PixVerse. Run PixVerse V6 on APImage. Social clips from 1 to 15 seconds at 360p to 1080p, eight aspect ratios, anime, 3D, clay, comic and cyberpunk styles, optional audio. From 21 credits per 5 seconds. - [Seedance 2.0 Fast](https://apimage.org/models/seedance-2-0-fast): ByteDance. Run ByteDance Seedance 2.0 Fast on APImage. Seedance motion with native audio at 480p or 720p, 4 to 15 seconds, 9 reference images, clips and audio. From 91 credits per 5 seconds. - [Seedance 2.0 Mini](https://apimage.org/models/seedance-2-0-mini): ByteDance. Run ByteDance Seedance 2.0 Mini on APImage. The cheapest Seedance, with native audio at 480p or 720p, 4 to 15 seconds, 9 reference images, clips and audio. From 61 credits per 5 seconds. - [GPT Image 2](https://apimage.org/models/gpt-image-2): OpenAI. Run OpenAI GPT Image 2 on APImage. Best in class text rendering including non Latin scripts, up to 4K, reasoning before generation. From 3 credits per image, on every plan. - [GPT Image 2.5 Flare](https://apimage.org/models/gpt-image-2-5-flare): OpenAI. Run OpenAI GPT Image 2.5 Flare on APImage. Higher quality than GPT Image 2 at 50 percent lower latency, with stronger reference photos and precise localized editing. - [GPT Image 2.5 Sunburst](https://apimage.org/models/gpt-image-2-5-sunburst): OpenAI. Run OpenAI GPT Image 2.5 Sunburst on APImage. The premium 2.5 model, built for production ready campaign creative and polished product imagery with precise multi turn editing. - [FLUX 3 Image](https://apimage.org/models/flux-3-image): Black Forest Labs. Run Black Forest Labs FLUX 3 Image on APImage. Text to image, precise edits and up to 10 reference images per generation, with optional web grounding. From 4 credits per image, on every plan. - [Nano Banana Pro](https://apimage.org/models/nano-banana-pro): Google. Run Google Nano Banana Pro on APImage. Photoreal images, legible text and precise multi-image edits at 1K, 2K or 4K, 8 reference images, optional web search. From 25 credits per image. - [Nano Banana 2](https://apimage.org/models/nano-banana-2): Google. Run Google Nano Banana 2 on APImage. Fast image generation and editing from 0.5K to 4K, 8 references, 14 aspect ratios including 4:1 and 8:1, optional web search. From 10 credits per image. - [Nano Banana](https://apimage.org/models/nano-banana): Google. Run Google Nano Banana (Gemini 2.5 Flash Image) on APImage. Quick, cheap image generation and editing with up to 4 references, nine aspect ratios. 7 credits per image. - [Flux 2 Max](https://apimage.org/models/flux-2-max): Black Forest Labs. Run FLUX 2 Max from Black Forest Labs on APImage. Maximum detail and prompt adherence, up to 4K, 4 reference images. 6 credits at HD, 8 at Full HD, 13 at 4K. - [Flux 2 Pro](https://apimage.org/models/flux-2-pro): Black Forest Labs. Run FLUX 2 Pro from Black Forest Labs on APImage. High quality image generation and editing with 4 reference images, from 3 credits per image. - [Flux 2 Klein 9B](https://apimage.org/models/flux-2-klein-9b): Black Forest Labs. Run FLUX 2 Klein 9B from Black Forest Labs on APImage. Fast, reliable image generation with 4 references, 2 credits per image, on every plan including free. - [Seedream 5 Lite](https://apimage.org/models/seedream-5-lite): ByteDance. Run ByteDance Seedream 5 Lite on APImage. Dense layouts and text, 8 reference images, 2K or 4K output, nine aspect ratios. 6 credits per image. - [Seedream 4.5](https://apimage.org/models/seedream-4-5): ByteDance. Run ByteDance Seedream 4.5 on APImage. Crisp 2K and 4K images and multi-reference edits with up to 8 references, nine aspect ratios. 7 credits per image. - [Ideogram 3](https://apimage.org/models/ideogram-3): Ideogram. Run Ideogram 3 on APImage. Posters, logos and typography with 62 looks, 8 colour palettes, style presets, reference remixing and negative prompts. From 5 credits per image. - [Recraft V4.1](https://apimage.org/models/recraft-v4-1): Recraft. Run Recraft V4.1 on APImage. Illustration, brand graphics and clean design compositions in five aspect ratios, standard or high-detail Pro. From 6 credits per image. - [Photo Relight](https://apimage.org/models/photo-relight): Google. Relight any photo on APImage. Set the light's direction, height, soft or hard quality, brightness and colour; everything else in the photo stays the same. Runs on Nano Banana 2. 14 credits per image. - [Photo Angles](https://apimage.org/models/photo-angles): Alibaba. Get any photo from another camera angle on APImage. Orbit 360 degrees, set the camera height from low angle to bird's-eye, and zoom from wide to close-up. Runs on Qwen Image Edit. 6 credits per angle. - [Kling Avatar 2](https://apimage.org/models/kling-avatar-2): Kuaishou. Make a photo talk with Kling Avatar 2 on APImage. A portrait plus a recording or a typed script becomes up to 60 seconds of lip-synced talking video. 94 credits per 10 seconds. - [Kling Avatar 2 Pro](https://apimage.org/models/kling-avatar-2-pro): Kuaishou. Run Kling Avatar 2 Pro on APImage. Higher-fidelity talking avatar from one photo and a voice, up to 60 seconds, with a recording or a typed script. 192 credits per 10 seconds. - [Kling Lip Sync](https://apimage.org/models/kling-lip-sync): Kuaishou. Run Kling Lip Sync on APImage. Make the person in a 2 to 10 second clip say new audio, at 720p or 1080p. 12 credits per 5 seconds. - [VEED Lip Sync](https://apimage.org/models/veed-lip-sync): VEED. Run VEED Lip Sync on APImage. Make the person in a video up to 60 seconds long say new audio. 117 credits per 10 seconds. - [ElevenLabs v3](https://apimage.org/models/elevenlabs-v3): ElevenLabs. Run ElevenLabs v3 text to speech on APImage. The most expressive voices, directed with audio tags like [whispers] and [laughs], 21 voices, 29 languages. 17 credits per 1,000 characters. - [ElevenLabs Multilingual v2](https://apimage.org/models/elevenlabs-multilingual-v2): ElevenLabs. Run ElevenLabs Multilingual v2 on APImage. Stable, natural text to speech in 29 languages with 21 voices and control over speed, stability and style. 17 credits per 1,000 characters. - [ElevenLabs Turbo v2.5](https://apimage.org/models/elevenlabs-turbo-v2-5): ElevenLabs. Run ElevenLabs Turbo v2.5 on APImage. Fast, low-cost text to speech for drafts and long scripts, 21 voices, 29 languages. 9 credits per 1,000 characters. - [Chatterbox](https://apimage.org/models/chatterbox): Resemble AI. Clone your voice with Chatterbox on APImage. Speak any script in a voice cloned from a 3 to 60 second sample, with control over expressiveness and pace. 5 credits per 1,000 characters. - [ElevenLabs Voice Changer](https://apimage.org/models/elevenlabs-voice-changer): ElevenLabs. Run ElevenLabs Voice Changer on APImage. Re-voice a recording up to 5 minutes long in one of 21 voices, keeping its timing, emotion and delivery. 50 credits per minute. - [ElevenLabs Dubbing](https://apimage.org/models/elevenlabs-dubbing): ElevenLabs. Dub a video with ElevenLabs Dubbing on APImage. Translate speech into 29 languages in the original speakers' voices, up to 5 minutes and 5 speakers. 100 credits per minute. - [MiniMax Music 2.6](https://apimage.org/models/minimax-music-2-6): MiniMax. Run MiniMax Music 2.6 on APImage. Full songs with vocals from a style description, with your lyrics or its own, or instrumental. 25 credits per song. - [Eleven Music](https://apimage.org/models/eleven-music): ElevenLabs. Run Eleven Music on APImage. Studio-grade music at the exact length you need, from 10 seconds to 5 minutes, with vocals or instrumental. 100 credits per minute. - [Lyria 3](https://apimage.org/models/lyria-3): Google. Run Google Lyria 3 on APImage. 30 second music clips with vocals from a description, quick and cheap for trying ideas. 7 credits per clip. - [Eleven Sound Effects](https://apimage.org/models/eleven-sound-effects): ElevenLabs. Run Eleven Sound Effects on APImage. Any sound from a description, from 1 to 22 seconds, with seamless loops for ambience. 1 to 8 credits per sound. - [Kling Video to Audio](https://apimage.org/models/kling-video-to-audio): Kuaishou. Run Kling Video to Audio on APImage. Add matching sound effects and music to a silent 3 to 20 second clip, with optional direction and ASMR mode. 6 credits per clip. - [VEED Clean Audio](https://apimage.org/models/veed-clean-audio): VEED. Run VEED Clean Audio on APImage. Remove background noise and get studio-quality speech from audio or video up to 20 minutes, with loudness targeting. 3 credits per minute. - [Kling 3.0 Motion Control](https://apimage.org/models/kling-3-0-motion-control): Kuaishou. Run Kling 3.0 Motion Control on APImage. Your character performs every move of a reference clip up to 30 seconds, with optional Face Lock and camera copy, at 720p or 1080p. From 210 credits per 10 seconds. - [Kling 2.6 Motion Control](https://apimage.org/models/kling-2-6-motion-control): Kuaishou. Run Kling 2.6 Motion Control on APImage. Copy a clip's moves or camera move onto your character's photo, up to 30 seconds at 720p or 1080p. From 117 credits per 10 seconds. - [DreamActor v2](https://apimage.org/models/dreamactor-v2): ByteDance. Run ByteDance DreamActor v2 on APImage. Motion Control at the lowest price, best with cartoon, non-human and multiple characters, clips up to 30 seconds. 84 credits per 10 seconds. - [Kling O3 Edit](https://apimage.org/models/kling-o3-edit): Kuaishou. Run Kling O3 Edit on APImage. Change the outfit, objects, look or lead in a 3 to 15 second clip with a prompt and up to 4 references, keeping every move and the original audio. From 234 credits per 10 seconds. - [PixVerse Swap](https://apimage.org/models/pixverse-swap): PixVerse. Run PixVerse Swap on APImage. Replace the person, an object or the background in a clip up to 10 seconds, at 540p or 720p, keeping the sound. From 50 credits per 10 seconds. - [MiniMax H3 Max](https://apimage.org/models/minimax-h3-max): MiniMax. Run MiniMax H3 Max on APImage. Recast a video of 5 to 30 seconds with up to four characters, keeping the scene, the moves and the sound, at 768p or 1080p. From 500 credits per 10 seconds. - [Wan 2.7 Edit](https://apimage.org/models/wan-2-7-edit): Alibaba. Run Alibaba Wan 2.7 Edit on APImage. Edit or recast a 2 to 10 second clip with a prompt and one photo, keeping the sound, at 720p or 1080p. From 167 credits per 10 seconds. - [PixVerse V6 Extend](https://apimage.org/models/pixverse-v6-extend): PixVerse. Extend a video with PixVerse V6 Extend on APImage. Continue any clip up to 30 seconds for 1 to 15 more seconds from its last frame, at 360p to 1080p, with optional sound. From 21 credits per 5 seconds added. - [Veo 3.1 Fast Extend](https://apimage.org/models/veo-3-1-fast-extend): Google. Extend a video with Veo 3.1 Fast Extend on APImage. Add 7 seconds in Veo quality to any clip up to 30 seconds, with optional sound. 117 credits per 7 seconds. - [LTX 2.3 Reframe](https://apimage.org/models/ltx-2-3-reframe): Lightricks. Reframe a video with Lightricks LTX 2.3 Reframe on APImage. Change a clip's aspect ratio to 9:16, 16:9, 1:1, 4:5 or 5:4 and fill the new areas, up to 20 seconds. From 167 credits per 10 seconds. - [ByteDance Video Upscaler](https://apimage.org/models/bytedance-video-upscaler): ByteDance. Upscale a video with the ByteDance Video Upscaler on APImage. Sharpen clips up to 60 seconds to 1080p, 2K or 4K, optionally at 60 fps, with presets for AI, phone and old footage. From 12 credits per 10 seconds. - [Bria Video Background Removal](https://apimage.org/models/bria-video-background-removal): Bria. Remove a video background with Bria on APImage. Cut the subject out of a clip up to 60 seconds onto transparency, green, black, white, blue or grey, keeping the audio. 84 credits per 10 seconds. - [Depth Anything Video](https://apimage.org/models/depth-anything-video): Depth Anything. Turn a video into a depth map with Depth Anything Video on APImage. Clips up to 60 seconds, grayscale or four colour maps, optional side-by-side view. 67 credits per 10 seconds. ## Model comparisons - [Veo 3.1 vs Kling 3.0](https://apimage.org/compare/veo-3-1-vs-kling-3-0): Veo 3.1 vs Kling 3.0 compared: resolution, length, audio, formats and price per clip. Veo wins on realism, dialogue and 4K; Kling on length, multi-shot and price. Run both on one APImage plan. - [Seedance 2.0 vs Veo 3.1](https://apimage.org/compare/seedance-2-0-vs-veo-3-1): Seedance 2.0 vs Veo 3.1 compared: references, length, resolution, audio and price. Seedance wins on control, length and formats; Veo on realism and price at 1080p and 4K. Run both on APImage. - [Seedance 2.0 vs Kling 3.0](https://apimage.org/compare/seedance-2-0-vs-kling-3-0): Seedance 2.0 vs Kling 3.0 compared: resolution, references, formats, audio, multi-shot and price per clip. Kling is a quarter of the price at 1080p; Seedance reaches 4K and takes references. Use both on APImage. - [FLUX 3 Video vs Veo 3.1](https://apimage.org/compare/flux-3-video-vs-veo-3-1): FLUX 3 Video vs Veo 3.1 compared: length, resolution, audio, keyframes, formats and price. FLUX 3 Video includes sound, runs to 20 seconds and costs less; Veo 3.1 reaches 4K. Use both on APImage. - [Seedance 2.5 vs Seedance 2.0](https://apimage.org/compare/seedance-2-5-vs-seedance-2-0): Seedance 2.5 vs Seedance 2.0 compared: length, references, resolution, draft mode and price. 2.5 runs to 30 seconds with 30 references; 2.0 reaches 4K and costs less per second. Both on APImage. - [Wan 3.0 vs Kling 3.0](https://apimage.org/compare/wan-3-0-vs-kling-3-0): Wan 3.0 vs Kling 3.0 compared: length, references, editing, frames, audio and price. Wan runs to 30 seconds with image, clip and audio references; Kling has frames, multi-shot and cheaper 1080p. Both on APImage. - [Kling 4.0 vs Kling 3.0](https://apimage.org/compare/kling-4-0-vs-kling-3-0): Kling 4.0 vs Kling 3.0 compared: length, references, audio, formats, controls and price per clip. Kling 4.0 wins on length, references, 21:9 and 720p price; Kling 3.0 on 1080p price and control. Run both on one APImage plan. - [Kling 4.0 vs Seedance 2.5](https://apimage.org/compare/kling-4-0-vs-seedance-2-5): Kling 4.0 vs Seedance 2.5 compared: two 30 second models on references, formats, drafts and price per clip. Kling 4.0 costs a fifth at 1080p with 10 references; Seedance 2.5 takes 30, plus clips and audio. Run both on APImage. - [Kling 4.0 vs Veo 3.1](https://apimage.org/compare/kling-4-0-vs-veo-3-1): Kling 4.0 vs Veo 3.1 compared: length, resolution, formats, audio and price per clip. Kling 4.0 wins on length, formats and price; Veo 3.1 on realism, 4K and silent clips. Run both on one APImage plan. - [GPT Image 2.5 Flare vs Nano Banana Pro](https://apimage.org/compare/gpt-image-2-5-flare-vs-nano-banana-pro): GPT Image 2.5 Flare vs Nano Banana Pro compared: text rendering, references, editing, grounding, plans and price. An HD Flare image costs 3 to 17 credits, Nano Banana Pro 25 to 50. Run both on APImage. - [Nano Banana 2 vs Nano Banana Pro](https://apimage.org/compare/nano-banana-2-vs-nano-banana-pro): Nano Banana 2 vs Nano Banana Pro compared: price per size, resolution, aspect ratios, references, grounding and quality. Nano Banana 2 is cheaper at every size; Pro is Google's best. Both on APImage. - [FLUX 3 Image vs Nano Banana Pro](https://apimage.org/compare/flux-3-image-vs-nano-banana-pro): FLUX 3 Image vs Nano Banana Pro compared: references, layout control, grounding, formats, plans and price. FLUX 3 Image takes 10 references for 4 to 8 credits; Nano Banana Pro is Google's photoreal model. Both on APImage. - [MiniMax Music 2.6 vs Eleven Music](https://apimage.org/compare/minimax-music-2-6-vs-eleven-music): MiniMax Music 2.6 vs Eleven Music compared: lyrics, length control, vocals and price. MiniMax writes full songs to your lyrics for 25 credits; Eleven Music cuts tracks to an exact length. Both on APImage. ## Developers - [Documentation](https://apimage.org/docs): REST API and MCP server overview - [API reference](https://apimage.org/docs/api-reference): Every endpoint, parameter and credit cost - [MCP server](https://apimage.org/docs/mcp): All 20 MCP tools and how to connect a client - [MCP and CLI](https://apimage.org/mcp): Connecting Claude, Cursor, ChatGPT and Codex - [Integrations](https://apimage.org/integrations): Zapier, Pipedream, n8n and the REST API ## Use cases: Creators & faceless channels Post daily, stay off camera, let volume do the work. - [Faceless YouTube Automation](https://apimage.org/use-cases/faceless-youtube-automation): Go viral on a channel that never shows your face. - [Faceless TikTok Channels](https://apimage.org/use-cases/faceless-tiktok-channels): Post ten a day, stay anonymous, go viral on the two that land. - [YouTube Shorts Automation](https://apimage.org/use-cases/youtube-shorts-automation): Fill the Shorts schedule and let volume find the one that goes viral. - [Instagram Reels Growth](https://apimage.org/use-cases/instagram-reels-growth): Reels that look shot on a phone, because that is what goes viral. - [AI Influencer & Virtual Creator](https://apimage.org/use-cases/ai-influencer): Build a creator who fronts every video and never misses a viral window. - [Reddit Story & Narration Videos](https://apimage.org/use-cases/reddit-story-videos): The narrated story format that goes viral, on visuals nobody else is using. - [Affiliate Marketing Videos](https://apimage.org/use-cases/affiliate-marketing-videos): A fresh angle every day until one goes viral and the link pays. - [TikTok Shop Affiliate Videos](https://apimage.org/use-cases/tiktok-shop-affiliate): Commission content for the whole catalogue, posted until one goes viral. ## Use cases: Ecommerce & brands Product content the day the stock lands. - [Shopify Product Videos](https://apimage.org/use-cases/shopify-product-videos): Sell more from the PDP with video the day the stock lands. - [Dropshipping Ad Creatives](https://apimage.org/use-cases/dropshipping-ads): Test twenty creatives, scale the one that sells more. - [Amazon Listing Images & Video](https://apimage.org/use-cases/amazon-listing-content): Fill all nine gallery slots and sell more from the same listing. - [Print on Demand Mockups](https://apimage.org/use-cases/print-on-demand): Flat template mockups do not sell. Products that look real do. - [Etsy Product Photography](https://apimage.org/use-cases/etsy-product-photos): Handmade sells more when it does not look photographed on a sofa. - [Fashion & Apparel Content](https://apimage.org/use-cases/fashion-and-apparel): Every colourway on every body type, so more of them sell. - [Beauty & Skincare Creatives](https://apimage.org/use-cases/beauty-and-skincare): Texture, application and packaging, shot the way beauty sells. ## Use cases: Business & agencies Client work and pipeline without a production budget. - [Agencies & Freelancers](https://apimage.org/use-cases/agencies-and-freelancers): Take on more retainers and sell more client work without hiring. - [SaaS & App Marketing Video](https://apimage.org/use-cases/saas-and-app-marketing): Launch creative that turns trials into paid seats. - [Real Estate Listing Media](https://apimage.org/use-cases/real-estate-listings): Staged interiors sell listings faster than empty rooms do. - [Local Business Social Content](https://apimage.org/use-cases/local-business-social): A month of posts that fill the salon, the gym and the restaurant. - [Coaches & Course Creators](https://apimage.org/use-cases/coaches-and-course-creators): Daily presence that sells the course, without daily filming. ## AI tools These need an account, because each run costs GPU time. The free plan includes 3 watermarked images so output quality can be judged before paying. - [All AI tools](https://apimage.org/tools): The full catalog ### Video - [TikTok Video Maker](https://apimage.org/tools/tiktok-video-maker): Turn a product photo and one line into a native 9:16 clip built to stop the scroll. - [YouTube Shorts Maker](https://apimage.org/tools/youtube-shorts-maker): Fill the upload schedule with vertical Shorts, without filming any of them. - [Instagram Reels Maker](https://apimage.org/tools/instagram-reels-maker): Creator-style Reels that look shot on a phone, because that is what performs. - [UGC Ad Video Maker](https://apimage.org/tools/ugc-video-maker): The creator-led ad format that converts, minus the creator, the phone and the shoot. - [Product Video Maker](https://apimage.org/tools/product-video-maker): One flat product photo in. A lit, moving, campaign-grade clip out. - [Talking Avatar](https://apimage.org/tools/talking-avatar): Give any face your audio and it delivers the script, up to 30 seconds. - [AI Influencer Generator](https://apimage.org/tools/ai-influencer-generator): Create a creator who never reschedules and fronts every video you post. - [YouTube Thumbnail Maker](https://apimage.org/tools/youtube-thumbnail-maker): Click-worthy covers with text that renders cleanly, not as melted letters. ### Generation - [Image Generator](https://apimage.org/tools/image-generator): Generate stunning images from text prompts using AI. - [AI Backgrounds](https://apimage.org/tools/ai-backgrounds): Create professional backgrounds for any subject. - [AI Templates](https://apimage.org/tools/ai-templates): Use pre-made AI templates to jumpstart your creative process. - [AI Posters](https://apimage.org/tools/ai-posters): Design eye-catching posters from simple prompts. ### Editing - [Edit / Inpaint Images](https://apimage.org/tools/edit-inpaint): Select and replace or remove any part of an image with context-aware AI. - [Remove Background](https://apimage.org/tools/remove-background): Remove backgrounds instantly for clean, transparent cutouts. - [Replace Background](https://apimage.org/tools/replace-background): Swap the background of an image with a new AI-generated setting. - [Remove Watermark](https://apimage.org/tools/remove-watermark): Clean up images by intelligently removing watermarks and text. - [AI Editing](https://apimage.org/tools/ai-editing): Advanced AI tools to refine and modify your images seamlessly. - [Short Photo Editor](https://apimage.org/tools/short-photo-editor): Quickly enhance and edit photos for social media. ### E-commerce - [Virtual Try-On Accessories](https://apimage.org/tools/virtual-try-on-accessories): See how accessories look on models with AI precision. - [Virtual Try-On Shoes](https://apimage.org/tools/virtual-try-on-shoes): Visualize sneakers and shoes on different feet. - [Virtual Try-On](https://apimage.org/tools/virtual-try-on): Let customers visualize clothing on varied body types. - [Fashion Studio](https://apimage.org/tools/fashion-studio): Design and visualize fashion layouts instantly. - [Fashion Studio for Shoes](https://apimage.org/tools/fashion-studio-shoes): Dedicated fashion workflows for footwear visuals. - [AI Model Photo](https://apimage.org/tools/ai-model-photo): Generate diverse human models wearing your products. - [Product Try-On](https://apimage.org/tools/product-try-on): Showcase physical products in realistic environments. - [AI Product Image Design](https://apimage.org/tools/ai-product-image-design): Design compelling product imagery for ads and listings. - [Real Estate Maker](https://apimage.org/tools/real-estate-maker): Enhance and stage interior and exterior property photos. ### Utilities - [Relight Image / Object](https://apimage.org/tools/relight-image): Instantly change the lighting direction and color on any image. - [Image Translator](https://apimage.org/tools/image-translator): Translate text within images while keeping original styling. - [AI Shadows](https://apimage.org/tools/ai-shadows): Add realistic drop shadows and reflections to flat objects. - [Style Clone](https://apimage.org/tools/style-clone): Extract the artistic style from one image and apply it to another. - [Image Enhancer](https://apimage.org/tools/image-enhancer): Automatically fix lighting, colors, and sharpness. - [Image Upscaler](https://apimage.org/tools/image-upscaler): Enhance image resolution up to 4× with AI super-resolution. ## Free browser tools 12 utilities that run entirely in the visitor's browser on the Canvas API. No account, no upload, no trial limit. They are free because they cost nothing to serve: the work happens on the visitor's own device and no file is ever sent to a server. - [Free tools section](https://apimage.org/tools#free): All 12 browser tools, on the main tool page - [YouTube Transcript to Text & SRT](https://apimage.org/tools/youtube-transcript-to-text): Paste a video's transcript from YouTube: clean text, clickable timestamps that play the video, TXT and SRT. - [Instagram Comment-to-DM Generator](https://apimage.org/tools/instagram-comment-to-dm-generator): Write the keyword and both DMs, check them against Instagram's limits, see the chat. - [Social Media Image Resizer](https://apimage.org/tools/social-image-resizer): Resize one image to every platform size at once. Feed, story, thumbnail, banner, profile. - [Aspect Ratio Cropper](https://apimage.org/tools/image-cropper): Crop to 9:16, 1:1, 4:5 or 16:9 with a draggable frame. Free, no upload, no account. - [TikTok & Reels Safe Zone Checker](https://apimage.org/tools/safe-zone-checker): See exactly where captions, buttons and the product card cover your video before you post. - [Instagram Carousel Splitter](https://apimage.org/tools/carousel-splitter): Cut one wide image into seamless carousel slides or a 3x3 grid. Free and instant. - [YouTube Thumbnail Preview Tester](https://apimage.org/tools/thumbnail-preview): See your thumbnail at the sizes people actually see it. Mobile feed, sidebar, search. - [Image Compressor](https://apimage.org/tools/image-compressor): Shrink file size without visibly wrecking the image. Compare before and after, then download. - [Image Converter](https://apimage.org/tools/image-converter): PNG, JPG and WEBP in any direction. Batch a whole folder, all in the browser. - [Color Palette Extractor](https://apimage.org/tools/color-palette-extractor): Pull the dominant colours out of any image as hex codes you can paste anywhere. - [Quote & Meme Card Maker](https://apimage.org/tools/text-on-image): Put readable text on any image. Quote cards, meme captions, thumbnail overlays. - [Watermark Adder](https://apimage.org/tools/add-watermark): Stamp your handle or logo on an image before you post it. Tiled or single, any opacity. ## Company and reference - [Use case index](https://apimage.org/use-cases): All 20 use cases in one place - [Blog](https://apimage.org/blog): Guides and tutorials for AI video, UGC ads and product content - [Prompt templates](https://apimage.org/prompt-templates): Copy-ready prompts with reference files and step-by-step workflows - [About](https://apimage.org/about): The team and the mission - [Support](https://apimage.org/resources/support): Help and support tickets - [Affiliate program](https://apimage.org/resources/affiliate): Affiliate program terms and commission - [Imprint](https://apimage.org/legal/imprint): Legal entity information (German Impressum) - [Privacy policy](https://apimage.org/legal/privacy): How data is handled - [Terms of service](https://apimage.org/legal/terms): Terms of use - [Acceptable use policy](https://apimage.org/legal/acceptable-use): Appendix A to the terms: allowed and forbidden uses, limits, moderation ## Contact - Support: ask@support.apimage.org - Sign up: https://app.apimage.org/signup - X/Twitter: https://twitter.com/apimageorg - LinkedIn: https://linkedin.com/company/apimageorg - Sitemap: https://apimage.org/sitemap.xml