Generates 5-15 second 480p or 768p video clips from a text prompt, first-frame image, or up to 12 image, video, and audio references, with synchronized dialogue, ambience, and sound effects.
# HeyGen Video 1.0 High-fidelity video generation with synchronized audio. ## Overview HeyGen Video 1.0 represents a significant leap in generative video technology, specifically engineered for marketing professionals and content creators who require high-quality, synchronized audiovisual output. Unlike standard video models that focus solely on visual motion, this model integrates dialogue, ambient sound, and sound effects directly into the generation process. By eliminating the need for separate lip-syncing or post-production audio layering, it provides a streamlined workflow for creating professional-grade video clips. The model is capable of generating 5-15 second clips in 480p or 768p resolutions. It offers extreme flexibility in input methods, allowing users to guide the generation using text prompts, first-frame images, or a combination of up to 12 reference assets including images, videos, and audio files. This makes it an ideal tool for projects where brand consistency and precise character movement are paramount. Whether you are creating a presenter-led advertisement or a dynamic product showcase, HeyGen Video 1.0 ensures that the visual output is perfectly aligned with the intended audio narrative. ## Use Cases * **Corporate Training Modules:** Generate consistent presenter-led videos that explain complex topics with perfectly synchronized speech. * **Social Media Ad Creative:** Create short, punchy 15-second promotional clips that feature branded characters or spokespeople. * **Product Launch Teasers:** Use reference images of new products to generate high-quality, cinematic reveal videos with integrated ambient soundscapes. * **Personalized Marketing Campaigns:** Leverage reference audio and image inputs to create localized or personalized video messages at scale. * **Interactive Content:** Produce short, engaging clips for landing pages that require immediate visual and auditory impact. * **Virtual Influencer Content:** Animate digital avatars for consistent brand storytelling across multiple social platforms. ## Parameters This model offers a robust suite of parameters to fine-tune your output, ranging from basic resolution and duration settings to advanced reference-based inputs. Please refer to the Input Parameters section in the Pixloop interface to configure your seed, aspect ratio, and reference media assets to achieve the desired creative outcome. ## Tips for Best Results * **Prioritize First-Frame Images:** For character-based videos, providing a high-quality first-frame image significantly improves facial consistency and lip-sync accuracy. * **Leverage Prompt Enhancement:** Use the 'turbo' prompt enhancement setting for general tasks to ensure the model interprets your intent with optimal creative flair. * **Match Aspect Ratios:** When using reference videos or images, ensure your aspect ratio parameter matches the source media to avoid unwanted cropping or distortion. * **Balanced Reference Inputs:** While the model supports up to 12 references, start with 1-3 high-quality assets to maintain better control over the final output style. * **Optimize for Quality:** If you are generating content for high-end digital displays, use the 768p resolution setting combined with a higher outputQuality value to minimize compression artifacts. ## About HeyGen Video 1.0 is developed by the team at HeyGen and provided via the Runware infrastructure. It is a specialized model designed to bridge the gap between static generative AI and professional video production, focusing on the seamless integration of visual motion and synchronized audio. By removing the traditional friction of post-production lip-syncing, this model empowers marketing teams to iterate faster and produce more engaging, human-centric video content.
Model input reference derived from preset schema.
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
| seed | integer | No | null | Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range. |
| prompt | string | Yes | null | Text prompt describing elements to include in the generated output. |
| duration | integer | No | 5 | Length of the generated video in seconds. The total number of frames produced is determined by duration multiplied by the model's frame rate (fps). |
| resolution | select | No | "768p" | Resolution preset for the output. When used with input media, automatically matches the aspect ratio from the input. |
| aspect_ratio | select | No | "16:9" | Shape of the video. Converted to width and height for the chosen resolution. Ignored when a first-frame image is given (the video follows the image). |
| outputFormat | select | No | "MP4" | Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities. |
| numberResults | integer | No | 1 | Number of results to generate. Each result uses a different seed, producing variations of the same parameters. |
| outputQuality | integer | No | 95 | Compression quality of the output. Higher values preserve quality but increase file size. |
| reference_audios | audio | No | null | List of reference audios (UUID or URL). |
| reference_images | image | No | null | List of reference images (UUID, URL, Data URI, or Base64). |
| reference_videos | video | No | null | List of reference videos (UUID or URL). |
| first_frame_image | image | No | null | An array of frame-specific image inputs to guide video generation. Each item can be either a plain image input (UUID, URL, Data URI, or Base64) or an object that pairs an image with a target position in the video. |
| promptEnhancement | select | No | "turbo" | Level of automatic prompt rewriting applied before generation. |