HappyHorse 1.1 Reference-to-Video

HappyHorse 1.1 Reference-to-Video is the R2V member of a three-variant video generation family that also includes text-to-video and image-to-video modes, all reached through a single API endpoint where only the model parameter changes. Third-party aggregator listings describe the model as taking a text prompt together with one or more image references and producing a video clip, and they expose a generation form with a broad set of aspect ratios (including 16:9, 9:16, 1:1, 4:3, 3:4, 4:5, 5:4, 9:21, and 21:9), 720p or 1080p output quality, and a configurable duration between roughly 3 and 15 seconds. The variant is positioned alongside other contemporary video APIs in the same provider rail, suggesting a production-oriented role rather than a research preview.

In practical terms, the R2V variant is best suited to workflows where a specific visual reference needs to anchor the look of the output, such as brand-consistent product clips, character-driven shorts, or style-locked social content, because the reference-plus-prompt input gives more control than text-only generation while remaining simpler than multi-shot conditioning. EvoLink surfaces per-second billing in the $0.124–$0.166 range (about 8.39–11.25 credits per second), which lets teams size cost directly against clip length and choose 720p for drafts and 1080p for finals. Because all three HappyHorse 1.1 variants share the same endpoint and parameter-style API, applications can fall back between T2V, I2V, and R2V without re-engineering integration, which makes the model family a flexible building block for mixed video pipelines rather than a single-purpose tool.