Step 1
Upload or generate a frame
Upload your photo, or generate a photoreal first frame with GPT Image to Video.
AI image to video · multi-model · 25 free credits
Upload a still or generate a first frame, then bring it to life with realistic camera motion. Multi-model studio: Seedance 2.5, MiniMax H3, Gemini Omni, and GPT Image to Video. Start free with 25 credits.
Free = 25 credits, personal/trial (pricing FAQ). Paid unlocks commercial + watermark-free. Not unlimited free.
Honesty first (same as Free / pricing)
Free is for personal / trial use with 25 credits and no credit card (per /pricing FAQ). Commercial rights + watermark-free exports unlock on paid plans only: Basic €19.99 · Pro €39.99 · API €59.99. We do not claim unlimited free or Free = commercial + no watermark. Details: /free-ai-video-generator.
Upload a still (photo, product shot, artwork) plus an optional motion prompt. The model returns a short MP4 with camera moves and/or subject motion. That is the head-term job this page covers.
It is distinct from text-to-video (scene from text only, no starting frame - see /create/ai-video-generator or /free-ai-video-generator) and from talking-photo (lip-sync focus - adjacent, not a substitute: /create/talking-photo-maker).
Step 1
Upload your photo, or generate a photoreal first frame with GPT Image to Video.
Step 2
Seedance / MiniMax H3 / Gemini Omni / GPT I2V. Pan, zoom, cinematic, or AI pick.
Step 3
9:16 / 1:1 / 16:9. Upgrade for commercial watermark-free per /pricing.
Tool shortcut: /create/image-to-video
No invented Elo scores. Pick the fit for the shot.
Social / UGC still to motion. Fast iteration for Reels and ads.
Seedance 2.5 landingHigher-fidelity / stereo when available. API id veo3 = MiniMax H3, not Google Veo.
Chat-style refine, then animate in Studio.
Gemini Omni landingGenerate a photoreal first frame, then animate. Model spoke (self-canonical).
GPT Image to Video pipelineOpen the create tool: /create/image-to-video
Practical guidance only - no fake benchmarks.
Product motion for Meta / TikTok from existing packshots.
Animate IG stills into 9:16 clips.
Film and agency boards before a shoot.
Illustration and concept art into living loops.
Cinemagraph-style hero clips for sites and decks.
Self-canonical pages. Cross-link, never steal each other's canonical.
| Job | Best NBV page |
|---|---|
| Animate my photo / product shot | This page + /create/image-to-video |
| Prompt to photoreal still to video | /gpt-image-to-video |
| Scene from text only | /create/ai-video-generator / /free-ai-video-generator |
| Story / narration | /create/story-to-video |
Same rules as /free-ai-video-generator and the /pricing FAQ: 25 credits, no card, personal/trial on Free. Paid = commercial rights + watermark-free. Do not copy create-page badges that imply Free = commercial + no watermark.
REST image_url to MP4: /developers/image-to-video-api/ · docs /docs/image-to-video.php. MCP studio credits: /mcp. Not the primary consumer CTA - register or the create tool first.
A tool that animates a still into a short video with camera/subject motion. Distinct from text-to-video and from talking-photo lip-sync.
Yes - upload any clear still and describe the motion.
Free = 25 credits, no card; personal/trial per /pricing. Not unlimited.
Free = personal/trial. Paid plans unlock commercial + watermark-free.
GPT spoke = OpenAI still generation + animate (/gpt-image-to-video). This hub = multi-model animate (incl. bring-your-own image). Both self-canonical.
Seedance 2.5, MiniMax H3 (veo3 = H3, not Google Veo), Gemini Omni, GPT Image to Video.
Up to 1080p; 9:16, 1:1, 16:9 (model-dependent); 4K upscale on Pro when available.
Yes - export 9:16.
Same page (EN). FR aliases 301 here.
/developers/image-to-video-api/ + /docs/image-to-video.php. MCP: /mcp.
25 credits, no card, personal/trial. Paid plans unlock commercial rights and watermark-free exports.