Cycling through all six. Tap any point to stop.

Measured on six things

AI video generation from photos and text prompts in a browser.

Speed

Desktop Lighthouse 96 but mobile 67 with a 5.7s LCP — the heavy hero image drags first paint on phones, where most creators actually are.

Screenshots

1 / 1

imagetovideo.io

About this app

imagetovideo.io helps creators make short AI videos using an uploaded photo or a written scene. Its browser interface provides a selection of video models, prompts for movement and camera direction, and video effects and templates. New users receive signup credits to try the service. Additional generation uses paid credits.

Key features

1Image-to-video generation with multiple selectable AI models including Kling, Pixverse, Wan and Veo
2Text-to-video generation from prompts alone, with customizable duration and style presets
3One-click video effect templates including trending AI Hug and Kiss effects
4Credit-based pricing with per-credit cost shown, failed generations refunded, and 100-day history

Target customer profile

Drawn from the product itself, not from a survey.

Demographic

Individual creators

Pain points

Turning still photos into moving scenes

Primary needs

Photo animation, model selection, and prompt controls.

Demographic

Marketing teams

Pain points

Developing visual ideas for promotional clips

Primary needs

Image-to-video and text-to-video workflows in a browser.

Demographic

People exploring AI video

Pain points

Trying animation ideas before purchasing more credits

Primary needs

Signup credits and video effect templates.

Scout Review

Written by AI from measured evidence, scored out of 100.

imagetovideo.io is a competent, clearly-scoped AI video tool with a real product behind the landing page: image-to-video, text-to-video, multiple named models (Kling, Pixverse, Wan, Veo), effect templates, and a credit-based pricing page that's unusually transparent about per-credit cost and refunds on failed generations. The measured evidence backs up the basics — 12/12 on security headers, 96/100 accessibility, 96/100 desktop performance. The problems are specific and fixable: mobile LCP of 5.7s, empty dark rectangles in the feature sections that look like broken images, a cramped hero input card, contrast failures on the hero text, and no docs or help center. Against RunwayML, this is a lighter, cheaper, model-aggregating alternative rather than a creative suite — which is a legitimate position, just not a differentiated one yet. The product works. The presentation is where the points are being left on the table.

Measure by measure

Usability

Clear 3-step flow and a detailed pricing FAQ, but no visible docs or help center, and the cramped hero input area makes the core action feel fiddly.

Design

Consistent dark theme and punchy white CTAs, but the hero prompt card is cramped and large empty dark rectangles in feature rows read as broken image containers.

Speed

Desktop Lighthouse 96 but mobile 67 with a 5.7s LCP — the heavy hero image drags first paint on phones, where most creators actually are.

Security

Full security header suite, HSTS, SPF and DMARC all verified — 12/12 on the probe. No published audits or bug bounty, which is normal at this stage.

Accessibility

Lighthouse a11y 96/100 with only contrast and accessible-name failures. Strong automated result, but no a11y statement or keyboard-support notes found.

Growth

Real problem, real wedge — multi-model access (Kling, Pixverse, Wan, Veo) plus effects templates — but the segment is crowded and the site names no niche it owns.

Improvement areas

Mobile performance: LCP of 5.7s driven by the heavy hero background image
Design polish: empty dark rectangles in feature sections read as broken or missing images
Documentation: no docs or help center — the FAQ is the only support surface
Positioning: no page names a specific target audience or niche the tool serves better

Design & Usability

The design language is coherent — dark theme, blue accent, high-contrast white pill CTAs — and the three-step 'How to Convert' section communicates the flow without a manual. But the visual audit flags real problems: large empty dark rectangles in the lower feature sections that read as broken or unloaded image containers, a hero prompt card that crams a textarea, two toggles, a model dropdown and an aspect selector into one small panel with weak hierarchy, and white text sitting directly on a busy fantasy background with no scrim. Usability is competent but thin: the homepage explains the flow, the pricing page has a genuinely useful FAQ (formats, generation time, ownership, refunds), and support email is listed. What's missing is a docs or help center link in the nav, and any free trial beyond signup credits. For a tool whose core action is 'type a prompt, pick a model, generate,' the input area being the most cramped element on the page is a real usability miss.

Speed & Security

Speed is a tale of two devices. Desktop Lighthouse lands at 96/100, which is genuinely good. Mobile lands at 67/100 with an LCP of 5.7s and FCP of 4.7s — the heavy hero background image is almost certainly the culprit, and mobile is where most creators will actually open this. TBT (27ms) and CLS (0.001) are both excellent, so the problem is specifically first-paint weight, not runtime jank. Security is the strongest measured dimension: the HTTP probe returned 12/12, with HTTPS enforced, HSTS, CSP, frame protection, X-Content-Type-Options, Referrer-Policy and Permissions-Policy all present, plus SPF and DMARC on the DNS side. That's a well-configured deployment. What's absent is the paperwork layer — no SOC 2, ISO, pen-test summary or bug bounty mentioned anywhere on the site. For a tool handling user-uploaded images and generating paid output, a short security page would cost little and close an obvious trust gap.

Accessibility & Growth

Accessibility scores 80 on the strength of a 96/100 Lighthouse audit, with only two failing checks: insufficient color contrast and visible text labels without matching accessible names. The visual audit independently confirms the contrast issue — white headline text sits directly over a busy fantasy hero image with no scrim. That's a fixable, specific problem, not a systemic one. There's no accessibility statement, no keyboard-navigation notes, and no multi-language support mentioned, which keeps this out of the top band. Growth is a judgment on the idea: image-to-video is a real, growing need, and the site's clearest wedge is model aggregation — it explicitly names Kling, Pixverse, Wan and Veo as available engines, plus one-click effect templates like AI Hug and Kiss. That's a genuine differentiator against single-model converters. The limit is that this is a reseller position: the underlying models are rented, the segment is crowded, and the site names no specific niche (wedding photographers? realtors? TikTok creators?) it serves better than anyone else. Traction signals are thin — one social profile listed — but that's context, not the score.

Conclusion

The honest read: this is a working tool with a solid technical foundation and a marketing site that undersells it. The security headers are better than most indie AI tools bother with. The accessibility audit is genuinely strong. The pricing page does something rare — it tells you exactly what a credit costs and what happens when a generation fails. What's holding the score back is polish and positioning: the empty rectangles, the cramped input, the 5.7s mobile LCP, and the absence of any page that says 'here's who this is for.' Fix the hero image weight, fill or remove the empty panels, add a scrim behind the headline, and write one page naming a specific audience — wedding photographers, real estate agents, whoever — and this jumps several points without a single new feature. The bones are good. The finish isn't there yet.

Suggestions for improvement

Compress or lazy-load the hero background image — mobile LCP of 5.7s is the single biggest speed fix available.
Fill or remove the large empty dark rectangles in the feature sections; they currently read as broken images.
Add a scrim or gradient behind the hero headline to fix the contrast failure flagged by both Lighthouse and the visual audit.
Publish a short docs or help center page and link it in the nav — the FAQ is doing too much work alone.

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	imagetovideo.io	RunwayML
Pricing model	Credit packs from $29.90 (1,000 credits) to $499.90 (35,000 credits); no subscription; failed generations refunded	Subscription-based tiers; consult Runway's own pricing page for current plans
Model selection	Aggregates multiple third-party models: Kling, Pixverse, Wan, Veo	Proprietary in-house models (Gen series); no third-party model aggregation
Image-to-video workflow	Upload image, add optional start/end frames and prompt, pick model and aspect ratio, generate	Image-to-video supported within a broader creative suite; consult Runway for current capabilities
Effect templates	One-click trending templates including AI Hug and AI Kiss effects	No equivalent one-click viral effect template library advertised
Free access	Signup credits for new users; no ongoing free tier advertised	Free tier historically offered with limited credits; check Runway's current terms
Security posture	12/12 security headers verified (HSTS, CSP, frame protection, SPF, DMARC); no published audits	Enterprise-oriented security posture; consult Runway's trust page for certifications