Annual Sale!Get 30% offClaim Now→

Image to Video AI Generator

Animate any still photo into a 4–15 second MP4 clip in your browser. Switch between 9 image-to-video models — Seedance 2.5, Seedance 2.0 Mini, Kling 2.1, Wan 2.5 and Seedance 1.0 — inside one workspace.

What is Image to Video AI on Flux1 AI?

Image to Video AI turns a static PNG, JPG or WebP photo into a short animated video clip guided by a text motion prompt. Instead of locking you into a single vendor model, this page combines 9 active video models in one form: ByteDance's Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini and Seedance 1.0; Kuaishou's Kling 2.1 Standard and Kling 2.1 Master; and Alibaba's Wan 2.5 and Wan 2.2.

Because every model interprets camera movement, physics, lighting and audio differently, switching models on the same input image is the fastest way to get a usable shot. Use Seedance 2.0 Mini (the default on this page, set to 480p for the lowest starting cost in the Seedance 2 line) or Kling 2.1 Standard (25 credits for 5 seconds) to test motion ideas, then step up to Seedance 2.5 for native synchronized sound and first-and-last-frame transitions, or Kling 2.1 Master for 1080p footage.

Everything runs in the browser with no GPU, local install or separate vendor accounts. You can drop in a photo from your device or animate an image you just made on Flux1 AI with FLUX.2 Pro or FLUX 1.1 Pro. Every model shows its exact credit cost before you submit, and failed generations are refunded automatically.

Models in this picker
9 active models (Seedance 2.5 / 2.0 / 1.0, Kling 2.1, Wan 2.5 / 2.2)
Default model
Seedance 2.0 Mini at 480p (64 credits for 4 s)
Input formats
PNG, JPG or WebP, up to 10 MB
Frame control
First frame on all models; first + last frame on Seedance 2.x
Clip length
4–15 s slider (Seedance 2.x) or 5 s / 10 s (Kling, Wan, Seedance 1.0)
Resolution
480p, 720p or 1080p depending on model
Native audio
Supported on Seedance 2.5, Seedance 2.0 family and Wan 2.5
Typical wait
About 1–3 minutes depending on model and queue
Starting cost
From 13 credits (Seedance 1.0 Lite, 5 s 480p)

What this Image to Video workspace does well

9 video models without switching tabs

Test the same starting frame across Seedance 2.5, Seedance 2.0, Kling 2.1 and Wan 2.5 from one dropdown. You keep your uploaded image and prompt intact while comparing motion smoothness, speed and credit cost.

First-frame and first-and-last-frame transitions

Every model animates from a single starting image. When you pick any Seedance 2.x model, you can toggle First & last frame and upload both endpoints so the model bridges one composition cleanly into the next.

Optional native audio in the same render

Seedance 2.5, Seedance 2.0 (Standard, Fast and Mini) and Wan 2.5 generate synchronized ambient sound, footsteps, dialogue and music alongside the video — no separate foley pass required.

Upfront credit calculator and automatic refunds

Changing model, duration or resolution updates the credit total on the submit button immediately. If an upstream provider times out or rejects a job, your credits are returned automatically and your finished clips stay saved under My Videos.

Which model should you pick inside the generator?

All four families below are available directly inside the picker above. Pick by budget, audio need and target resolution.

Seedance 2.0 Mini & Fast

Best starting point for audio + first/last frame

  • 4–15 second duration slider at 480p or 720p with native audio.
  • Seedance 2.0 Mini is the page default: 64 credits (4 s 480p) or 175 credits (5 s 720p).
  • Supports First & last frame mode for controlled transitions.

Seedance 2.5 & Seedance 2.0

Higher motion fidelity with synchronized sound

  • Flagship physics, camera choreography and dialogue sync up to 15 seconds.
  • 5 s at 720p costs 360 credits on Seedance 2.0 and 560 credits on Seedance 2.5.
  • For multi-image omni references or clips up to 30 s, open Seedance 2.5 Flash.

Kling 2.1 Standard & Master

Strong subject consistency and 1080p option

  • Kling 2.1 Standard delivers 720p 5 s clips for 25 credits (10 s for 70 credits).
  • Kling 2.1 Master upgrades to 1080p with tighter prompt following (120 credits for 5 s).
  • Best for product pans, portraits and clean camera pushes without audio.

Wan 2.5, Wan 2.2 & Seedance 1.0

Lowest credit cost per clip for fast iteration

  • Seedance 1.0 Lite starts at 13 credits for a 5 s 480p silent clip (25 at 720p, 100 for Pro 1080p).
  • Wan 2.5 costs 25 credits for 5 s (50 for 10 s) with synced audio.
  • Wan 2.2 runs at 720p for 25 credits (5 s) with a steady camera.

Need 4K output, video editing, or multi-reference video inputs? Flux1 AI also has dedicated workspaces for Gemini Omni 1.2 Flash, Wan 3.0, Grok Imagine 1.5 and Veo 3.

Image to Video AI Examples

Input image 1

Input image #1

Generated video #1

Input image 2

Input image #2

Generated video #2

How to convert an image to video

  1. 1

    Upload your starting frame

    Drop a PNG, JPG or WebP image (up to 10 MB) into the frame box, or paste a direct image URL. Clean, well-lit photos with a clear subject animate more reliably than heavily cropped or blurry screenshots.

  2. 2

    Choose a model and frame mode

    Keep the default Seedance 2.0 Mini or pick another model from the two-column selector. On Seedance 2.x models, switch to First & last frame if you want the clip to land on a specific ending image.

  3. 3

    Describe the motion, not the whole scene

    The model can already see what is in your photo. Spend your prompt on what moves, how the camera travels, the pacing and (on audio models) what sounds should accompany the shot.

  4. 4

    Set duration, resolution and generate

    Pick 480p, 720p or 1080p and your clip length, check the credit cost on the button, and click generate. Most clips finish in 1–3 minutes and save automatically to your My Videos library as MP4.

Image-to-video prompt tips

Because the source image already fixes the subject, wardrobe and palette, short motion-first prompts outperform long visual descriptions.

Vague promptMake this photo move cinematically, high quality 4K masterpiece.
Motion-focused promptSlow dolly-in toward the coffee cup as steam curls upward in morning sunlight. Keep the table and background steady.
Vague promptA woman in a blue coat standing on a street in Tokyo at night with neon lights.
Motion-focused promptShe turns her head toward the camera and smiles gently while wind moves her hair. Soft rain falls past the out-of-focus neon signs behind her.
Vague promptCamera moves around fast with crazy action and explosions.
Motion-focused promptSmooth 45-degree orbit to the right around the sneaker at eye level, studio rim light gliding across the mesh texture.
Vague promptAnimate the dog on the beach.
Motion-focused promptThe dog shakes water from its coat, then trots two steps toward the lens. Gentle ocean waves roll across the wet sand; sound of surf and paws on sand.

Compare image-to-video models on Flux1 AI

Every credit figure below is pulled from the same billing tables our generators charge. Pick a model inside this page or jump to a dedicated workspace.

Model5 s–6 s 720p clipMax lengthAudioBest for
Kling 2.1 Standard (here)25 credits (5 s)10 sNoLowest-cost 720p character & product motion
Wan 2.5 (here)25 credits (5 s)10 sYesBudget clips with synchronized sound
Seedance 2.0 Mini (default here)175 credits (5 s; 64 at 4 s 480p)15 sYesFirst + last frame control and longer audio clips
Gemini Omni 1.2 Flash35 credits (5 s)15 sYesFast 720p–4K clips, video edit & extend
Grok Imagine 1.548 credits (6 s)15 sYesExpressive motion with sound at low per-second cost
Wan 3.090 credits (5 s)30 sYesUp to 1080p and 30-second multi-shot clips
Seedance 2.5 Flash340 credits (5 s)30 sYesOmni reference (10 images, 5 clips, 5 audio), edit & extend

Image to Video hub vs. dedicated model pages

Use this page when you want to test one photo across multiple engines (Seedance 2.x, Kling 2.1, Wan 2.5) without re-uploading. Open a dedicated page like Seedance 2.5 Flash or Gemini Omni 1.2 Flash when you need multi-image reference inputs, video editing, or clip extension.

Draft at 480p or Kling 2.1 before rendering flagship clips

A 5 s 720p clip costs 25 credits on Kling 2.1 Standard or 80 credits on Seedance 2.0 Mini at 480p, versus 560 credits on Seedance 2.5 at 720p. Dial in your camera direction on a cheaper tier first, then re-run the winning prompt on your final model.

Image to Video AI credit pricing

Every model on Flux1 AI draws from one shared credit balance. No per-model add-ons — you only pay for the clip you generate.

Model & settingWhat you getCredits
Seedance 1.0 Lite (5 s, 480p)Lowest-cost silent image-to-video draft in this picker13 credits
Kling 2.1 Standard / Wan 2.5 (5 s)720p silent (Kling 2.1) or synced audio (Wan 2.5); 10 s is 70 / 5025 credits
Seedance 2.0 Mini (4 s, 480p)Default setting on this page; native audio + first/last frame64 credits
Seedance 2.0 Mini (5 s, 720p)HD clip with native audio and smooth camera motion175 credits
Kling 2.1 Master (5 s, 1080p)Full HD 1080p output with strong subject adherence120 credits
Seedance 2.0 Fast / 2.0 / 2.5 (5 s, 720p)Higher tiers with audio: 280 / 360 / 560 credits280–560 credits
Worked example: 10 five-second clips10 clips at 720p on Kling 2.1 Standard or Wan 2.5250 credits

How far a monthly plan goes

  • Basic · 300 credits/mo

    ≈ 12 Kling 2.1 clips (5 s) or 4 Seedance 2.0 Mini drafts (4 s 480p)

  • Pro · 800 credits/mo

    ≈ 32 Kling 2.1 clips (5 s) or 12 Seedance 2.0 Mini drafts (4 s 480p)

  • Max · 2000 credits/mo

    ≈ 80 Kling 2.1 clips (5 s) or 31 Seedance 2.0 Mini drafts (4 s 480p)

New accounts receive 10 starter credits to try image models. Video clips start at 13 credits, so generating videos requires a credit pack or subscription. Failed runs are always refunded.

Good to know before you generate

Free accounts stay on the default model

Switching away from the default Seedance 2.0 Mini model in the picker requires an active paid plan. On a paid plan, all 9 models in this workspace and every dedicated video page unlock with the same credit pool.

Starter credits do not cover a full video clip

Sign-up gives 10 free credits, while the lowest-cost image-to-video clip here is 13 credits (Seedance 1.0 Lite) and the default Seedance 2.0 Mini 4 s draft is 64 credits. You will need a plan or credit pack before running video jobs.

No multi-image reference or video editing on this form

This form accepts a starting frame (plus an optional ending frame on Seedance 2.x). If you want to mix multiple character/object reference images, edit an existing clip, or extend a video, use Seedance 2.5 Flash or Gemini Omni 1.2 Flash.

Real waits and content screening

Most clips finish in 1–3 minutes, though peak traffic on 1080p or 15-second renders can take longer. You can leave the tab open or check My Videos later. Explicit sexual content and non-consensual likenesses are blocked before billing.

Image to Video AI FAQ

Can I use Image to Video AI for free?

New Flux1 AI accounts receive 10 free credits, which is enough to try image models like FLUX 1.1 Pro or FLUX.2 Pro. Because video generation is far heavier to compute, the cheapest clip in this workspace starts at 13 credits and the default 4 s Seedance 2.0 Mini clip costs 64 credits, so generating video requires a subscription or credit pack. Failed jobs are refunded automatically.

Which video models are included on this page?

The model picker includes 9 active image-to-video engines: Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, Seedance 1.0 (Lite and Pro), Kling 2.1 Standard, Kling 2.1 Master, Wan 2.5 and Wan 2.2. You can also switch to dedicated model pages for Veo 3, Wan 3.0, Grok Imagine 1.5 and Gemini Omni 1.2 Flash using the same credit balance.

What image formats and sizes can I upload?

Upload PNG, JPG or WebP files up to 10 MB, or paste a public image URL. By default the output video follows your input image's aspect ratio; models that expose aspect-ratio controls (such as Seedance 2.x and Wan 2.2) also let you choose standard ratios like 16:9, 9:16 and 1:1.

How long can an image-to-video clip be?

Inside this generator, Seedance 2.x models support any length from 4 to 15 seconds via the duration slider, while Kling 2.1, Wan 2.5, Wan 2.2 and Seedance 1.0 generate 5-second or 10-second clips. If you need clips up to 30 seconds, open Wan 3.0 or Seedance 2.5 Flash.

Can I control both the start frame and the end frame?

Yes. Select any Seedance 2.x model (Seedance 2.5, 2.0, 2.0 Fast or 2.0 Mini) and switch the frame toggle from First frame to First & last frame. Upload your opening and closing images, describe how the transition should happen, and the model interpolates the motion between them.

Do generated videos include sound?

That depends on the model you choose. Seedance 2.5, Seedance 2.0 (Standard, Fast and Mini) and Wan 2.5 generate synchronized ambient audio, effects and speech natively. Kling 2.1, Wan 2.2 and Seedance 1.0 produce silent MP4 clips — you can add sound afterward with Add Audio to Video.

Why are videos exported without watermarks on paid plans?

Paid subscribers get watermark-free MP4 downloads, full access to all 9 models in the picker, private generations, and commercial usage rights under the Flux1 AI plans. Downloads are standard MP4 files ready for Premiere, CapCut, YouTube Shorts, TikTok and Instagram Reels.

Does Image to Video AI allow NSFW content?

No. Explicit sexual content and policy-violating imagery are screened before generation and blocked by both Flux1 AI and upstream video providers. Blocked or failed runs do not deduct credits.

Animate your photo now

Upload a still image above, pick from 9 video models, and download an MP4 clip in minutes.