MiniMax H3 H3 Image-to-Video: Image to Video Online
Bottom line
MiniMax's H3 animates a still with native audio, 4-15s, 768P/2K, from 60 credits, aspect follows the image. Upload a starting frame, choose MiniMax H3 H3 Image-to-Video in the Image to Video tool, and animate it on VdoBloom — the credit price appears before every generation.
MiniMax H3 Image-to-Video takes a still image and animates it while generating synchronized audio in the same pass — the same audio-native approach as its Text-to-Video sibling, applied to a supplied source frame instead of a blank prompt. Durations run 4 to 15 seconds in one-second steps, at 768P or 2K.
Unlike the text-to-video variant, aspect ratio isn't a separate choice here — the output follows the shape of your uploaded image, so there's one less setting to configure, but it also means cropping or padding your source to the right frame before uploading matters more than it does on models with independent aspect-ratio controls.
Pricing matches the text-to-video sibling exactly: 15 credits/second at 768P, 24 credits/second at 2K — so a 4-second 768P clip is 60 credits and a 15-second 2K clip is 360. It carries a MODERATE content badge and the family's 4.9 rating, and new accounts start with 10 free credits toward their first render.
What it costs on VdoBloom
Identical per-second pricing to the text-to-video model: 15 credits/second at 768P and 24 credits/second at 2K, for 4 to 15 seconds. A 5-second 768P animation is 75 credits; the same duration at 2K is 120. Cost is shown before you generate, and a failed render is refunded automatically. New accounts start with 10 free credits.
Image-to-video vs text-to-video H3: the aspect-ratio tradeoff
The two H3 variants share pricing and duration options, but only the text-to-video model lets you pick from six aspect ratios independently. The image-to-video model instead inherits whatever shape your source image is — simpler to use when you already have a correctly framed still, but it means getting the crop right before upload matters more here than on models with a separate aspect-ratio control.
Common problems and how to avoid them
The most common image-to-video complaint across models — motion that doesn't match the prompt — applies here too: describe one clear action rather than a paragraph of instructions, since H3 has to interpret both the still image and the prompt together. Because pricing is per-second, testing a 4-second 768P animation before committing to a full 15-second 2K render is the cheapest way to confirm the motion looks right before scaling up.
What creators use MiniMax H3 H3 Image-to-Video for
- Animating a photo or illustration into a scene with built-in ambience or sound effects, no separate audio pass.
- Product or key-art animation where the source image's exact framing needs to carry through untouched.
- Long-form 15-second animations of a single still where sound timing matters (dialogue, effects, music beats).
- Testing a short 4-second 768P animation cheaply before committing to a longer 2K render.
How to use MiniMax H3 H3 Image-to-Video on VdoBloom
- 1Open the Image to Video tool and select MiniMax H3 H3 Image-to-Video from the model picker.
- 2Upload your starting image and describe the motion you want in the prompt.
- 3Choose your duration (4–15s), quality (768P/2K), review the credit cost, and generate.
Frequently asked questions
How much does MiniMax H3 Image-to-Video cost?
Same rate as the text-to-video variant: 15 credits/second at 768P, 24 credits/second at 2K. A 4-second 768P clip is 60 credits; an 8-second 2K clip is 192 credits.
Does MiniMax H3 Image-to-Video let me pick the aspect ratio?
No — unlike the text-to-video variant's six aspect ratio options, the image-to-video output follows the shape of the image you upload. Crop or pad your source image to the frame you want before uploading.
Does MiniMax H3 Image-to-Video generate sound?
Yes, in every render — audio is generated together with the animation, the same audio-native design as the text-to-video model, applied to your uploaded image instead of a text-only prompt.
Is MiniMax H3 free?
New accounts start with 10 free signup credits, which doesn't cover the cheapest 60-credit render outright but lets you explore the tool before buying credits.
MiniMax H3 vs Seedance 2.0 for image animation — which is cheaper?
Seedance 2 (bytedance/seedance-2) runs 82 credits flat for a 5-second 720p clip with no built-in audio. MiniMax H3's 5-second 768P clip is 75 credits and includes generated sound — cheaper and with an extra capability, though H3's resolution ceiling (2K) and Seedance 2's (4K) differ.
Is MiniMax H3 censored?
MODERATE — VdoBloom's standard filtering tier, applied to both the uploaded image and the prompt. It's less strict than the STRICT badge used by Kling or Veo, but explicit material is still refused.
What's the longest clip MiniMax H3 Image-to-Video can produce?
15 seconds, in one-second increments from a 4-second minimum. At 768P that's 225 credits; at 2K it's 360 credits — matching the text-to-video variant's pricing exactly.
Is MiniMax open source?
No, it's a proprietary hosted API model from MiniMax — there's no self-hosted or open-weights release of Hailuo/H3.
Why did my MiniMax H3 image-to-video output come out the wrong shape?
Because this model has no independent aspect-ratio setting, the output inherits your source image's dimensions. If the result looks stretched or cropped wrong, the fix is re-uploading an image already cropped to the frame you want, not adjusting a setting in the generator.