Veo 3.1
Veo 3.1 builds on the capabilities of the previous model to enable longer and more versatile AI-generated videos. With this version, users can create multi-shot clips guided by multiple prompts, generate sequences from three reference images, and use frames in video workflows that transition between a start and end image, both with native, synchronized audio. The scene extension feature allows extension of a final second of a clip by up to a full minute of newly generated visuals and sound. Veo 3.1 supports editing of lighting and shadow parameters to improve realism and scene consistency, and offers advanced object removal that reconstructs backgrounds to remove unwanted items from generated footage. These enhancements make Veo 3.1 sharper in prompt-adherence, more cinematic in presentation, and broader in scale compared to shorter-clip models. Developers can access Veo 3.1 via the Gemini API or through the tool Flow, targeting professional video workflows.
Learn more about Veo 3.1
Seedance 2.5
Seedance 2.5 is ByteDance Seed’s new-generation video creation model for long-form storytelling, multimodal reference-based generation, and precise video editing. The model can generate high-quality 30-second audio-video clips in a single pass and supports multi-round extensions for creating longer videos with consistent characters, environments, pacing, and audiovisual style. Seedance 2.5 accepts up to 30 images, 10 video clips, and 10 audio clips as references, giving creators more control over subjects, scenes, motion, camera work, and creative direction. It improves transitions, visual consistency, audio-video synchronization, object textures, skin and eye details, lighting, color, and cinematic realism. The model also supports timestamp-level editing, green screen editing, camera perspective editing, clay render referencing, motion referencing, and reference-based editing.
Learn more about Seedance 2.5
MAI Image 2.7
MAI Image 2.7 is Microsoft AI’s image-editing model for making targeted changes to existing images while preserving their broader composition. It can restyle or relight a scene, move objects into new settings, revise text and symbols in images, and restore or enhance visual assets. It is aimed at creators and others who need to refine images without rebuilding an entire scene. Preview access is available through MAI Playground, Arena, and Artificial Analysis. The initial preview focuses on image editing, with text-to-image generation planned for a later release. Example uses include changing a product scene, modifying an image’s mood, and making selective edits to individual objects.
Learn more about MAI Image 2.7
Kling 3.0 Omni
Kling 3.0 Omni model is a generative video system designed to create imaginative videos from text prompts, images, or reference materials using advanced multimodal AI technology. It allows users to generate continuous video clips with flexible durations ranging from approximately 3 to 15 seconds, enabling short cinematic scenes that respond closely to prompt instructions. It supports prompt-based video generation as well as reference-based workflows, where users provide images or other visual elements to guide the subject, style, or composition of the generated scene. It improves prompt adherence and subject consistency, allowing characters, objects, and environments to remain stable throughout the generated clip while maintaining realistic motion and visual coherence. The Omni model also enhances reference-based generation so that characters or elements introduced through images remain recognizable across frames.
Learn more about Kling 3.0 Omni