HappyHorse 1.1 Reference-to-Video is the R2V member of a three-variant HappyHorse 1.1 family from Alibaba's Token Plan (China) that also includes text-to-video and image-to-video siblings. All three variants are reached through one shared API endpoint on third-party gateways such as EvoLink, with only the model parameter changing, which keeps integration simple when switching between generation modes. The Reference-to-Video variant accepts a text prompt together with reference images and returns a generated video, supporting 720p and 1080p output, durations from 3 to 15 seconds, and a wide spread of aspect ratios including 16:9, 9:16, 1:1, 4:3, 3:4, 4:5, 5:4, 9:21, and 21:9, so the same model can be reused across social clips, widescreen cinematics, and square or portrait formats.

On EvoLink, HappyHorse 1.1 is positioned as a stable, managed-access option aimed at production workloads that need dashboard billing, API key control, and predictable integration behaviour, with per-second credit pricing roughly in the $0.124–$0.166 (about 8.39–11.25 credits per second) band. Because the R2V endpoint shares billing and quotas with the wider HappyHorse 1.1 family, teams already using the text-to-video or image-to-video variants can add reference-driven generation without learning a new interface, which makes the model a natural fit for workflows where consistent character or product appearance across shots matters more than chasing the highest possible resolution.