Kling O1 Reference Video

Generate new videos that match the motion and camera style of your reference video.

"Based on @Video1, generate the next shot. Show @Element1 in the same style"

Input Video

@Video1

Generated Video

Generated

Elements Used

@Element1

Frontal

Describe your scene and generate a video in seconds

8,500+ videos generated this month

📄 About Kling O1 Reference Video

Kling O1 Reference Video is an advanced AI-powered video generation model designed to create new video shots guided by a reference video. Leveraging cutting-edge video-to-video technology, this model enables users to generate visually cohesive and stylistically consistent video sequences, making it an ideal solution for creative professionals seeking cinematic results. By analyzing the input reference video's motion, camera style, and overall cinematic language, Kling O1 Reference Video can produce fresh shots that seamlessly fit into existing footage or projects. The model accepts short reference videos (3-10 seconds, up to 200MB, in .mp4 or .mov formats) and enables detailed customization through prompts. Users can specify not only how the new shot should look but also introduce new characters or objects using up to four reference images or elements. Integration of @Video1, @Image1-4, and @Element1-4 in your prompt empowers highly precise control over the style, appearance, and composition of the generated video. This makes Kling O1 Reference Video exceptionally versatile for various video creation needs, from storyboarding and pre-visualization to marketing and content production. Key features include support for multiple aspect ratios (auto, 16:9 landscape, 9:16 portrait, 1:1 square), the ability to keep original audio, and flexible duration options (5 or 10 seconds). The intuitive input schema is optimized for both simplicity and creative depth, allowing users to upload videos, images, and elements, and to craft detailed prompts for nuanced results. Whether you want to introduce a new character, extend a narrative, or match the visual language of existing footage, Kling O1 Reference Video provides the tools needed for consistent, high-quality output. Ideal use cases span across industries: filmmakers can quickly generate new shots for storyboarding or pre-visualization; content creators and marketers can craft engaging video snippets for campaigns; animators can experiment with style and movement continuity; and educational or training teams can produce tailored video materials that maintain visual coherence. The platform's pay-as-you-go credit system provides flexibility for projects of all sizes without long-term commitments. Overall, Kling O1 Reference Video stands out for its ability to preserve cinematic integrity while offering granular creative control. Its robust feature set, user-friendly interface, and high-quality output make it a go-to solution for anyone aiming to elevate their video production workflow with AI.

✨ Key Features

Generates new video shots guided by reference footage, preserving motion and camera style for cinematic results.

Supports up to four total references, including videos, images, and custom elements (characters or objects).

Flexible aspect ratio options including auto-match, 16:9 landscape, 9:16 portrait, and 1:1 square.

Allows users to retain original audio from the input video for seamless integration.

Customizable prompts with detailed referencing using @Video1, @Image1-4, and @Element1-4 for precise output control.

Quick video generation with options for 5 or 10-second outputs, suitable for short-form content.

Easy-to-use interface optimized for both creative professionals and beginners.

💡 Use Cases

⚡Generating new cinematic shots for films, storyboards, or pre-visualizations.

⚡Creating marketing and social media video snippets that match a brand's visual style.

⚡Enhancing educational or training videos with consistent visuals or new character introductions.

⚡Animating new scenes for content creators based on existing footage.

⚡Producing unique video content for advertisements, product demos, or explainer videos.

⚡Rapid prototyping of video concepts for agencies and production teams.

⚡Extending or remixing user-generated video content with professional quality.

🎯 Best For

🎯 Filmmakers, video editors, marketers, content creators, animators, and creative professionals seeking cinematic AI-generated video shots.

👍 Pros

✓Preserves the cinematic language and style of reference footage for cohesive results.

✓Highly customizable with support for multiple input types and detailed prompts.

✓Flexible output options for duration and aspect ratio to suit various platforms.

✓Easy integration of new characters or objects using reference images.

✓Suitable for a wide range of creative and professional applications.

✓Fast turnaround time for video generation, enabling rapid iteration.

⚠️ Considerations

△Limited to short video durations (5 or 10 seconds).

△Maximum of four total references (elements plus images) per generation.

△Input video files are restricted to 3-10 seconds and 200MB in size.

△More advanced customization may require detailed prompt engineering.

📚 How to Use Kling O1 Reference Video

Prepare a reference video (3-10 seconds, .mp4 or .mov format, up to 200MB) and upload it to the platform.

Write a detailed prompt using @Video1 for the video, @Image1-4 for reference images, and @Element1-4 for characters or objects as needed.

Optionally, upload up to four reference images or elements to guide the style or include new visuals.

Select your desired video duration (5 or 10 seconds) and aspect ratio (auto, 16:9, 9:16, or 1:1).

Choose whether to keep the original audio from the input video by checking the appropriate box.

Submit your request and wait for the generated video shot, which is typically ready in 1-2 minutes.

💡 Pro Tips for Kling O1 Reference Video

★

Match Your Reference Video Quality The output quality heavily depends on your input. Use reference videos with stable camera work, clear subject motion, and good lighting. Avoid shaky handheld footage or poorly lit scenes. Videos shot at 1080p or higher with consistent frame rates produce the most cinematic results. If your reference is blurry or low-resolution, the generated output will inherit those limitations.

★

Leverage Element References for Character Consistency When introducing new characters using @Element1-4, provide multiple reference angles in addition to the frontal image. This helps the model understand the character's full appearance and maintains consistency across different poses. For character-driven narratives, consider Bytedance Dreamactor v2 which specializes in actor-specific motion transfer with even tighter character control.

★

Craft Specific Motion Descriptions Generic prompts like 'based on @Video1, generate the next shot' work, but detailed motion descriptions yield better results. Specify camera movements (dolly in, pan left, orbit around subject) and subject actions (walking forward, turning head, reaching out). The more precise your prompt, the more control you have over the final output while maintaining the reference style.

★

Use Auto Aspect Ratio First The 'auto' aspect ratio option matches your input video dimensions, which typically produces the most natural results since the model preserves the original framing context. Only switch to 16:9, 9:16, or 1:1 when you need platform-specific output. For social media content requiring vertical video, compare with Kling Video v3 Motion Control Pro which offers additional motion control parameters.

★

Keep Audio for Seamless Edits If you're generating shots to insert into existing footage, enable 'keep audio' to maintain audio continuity. This is especially useful for extending scenes or creating B-roll that matches your primary footage. For projects requiring custom soundtracks, leave audio off and add music in post-production or use JAI Music Clip Generator for AI-generated music sync.

★

Start With 5-Second Outputs Generate 5-second clips first to validate your prompt and reference setup before committing credits to 10-second outputs. This iterative approach saves credits and helps you refine motion, framing, and style choices. Once you nail the look, scale up to 10 seconds. For longer sequences, generate multiple 5-10 second clips and stitch them in your video editor.

Ready to try Kling O1 Reference Video?

Get 10 free credits — no credit card required

Start Free →

Frequently Asked Questions

The model analyzes your reference video to capture motion, camera style, and overall cinematic language. It then generates new shots that match these visual elements, ensuring a cohesive and professional result.

Reference videos must be in .mp4 or .mov format, 3-10 seconds long, and under 200MB. Reference images for style, appearance, or elements can be in any standard image format.

Yes, you can introduce new characters or objects by uploading frontal and reference images as elements. Use @Element1-4 in your prompt to specify how these should appear in the generated video.

Absolutely. You can choose to retain the original audio by selecting the 'keep audio' option during setup, allowing for seamless integration with your existing footage.

Pricing varies by model and is based on a pay-as-you-go credit system. This provides flexibility, so you only pay for what you use, without long-term commitments.

Credit costs vary by duration and complexity. A 5-second generation typically consumes fewer credits than a 10-second output. The exact credit amount is displayed before you submit each generation request on JAI Portal. Since pricing is pay-as-you-go, you only pay for what you generate—no subscription required. For budget planning, test with 5-second clips first, then scale to 10 seconds once you're satisfied with the output. If you're generating multiple variations, consider models like Kling Video v2.6 Motion Control Standard which may offer different credit rates for similar motion-based video generation.

Yes, all videos generated with paid credits on JAI Portal come with full commercial-use rights. You can use the output in advertisements, social media campaigns, client projects, YouTube content, product demos, and any other commercial application without additional licensing fees. This includes both the video content and any retained audio from your reference video (assuming you own rights to the original audio). Always ensure your input reference videos and images are either original content you own or properly licensed. For large-scale commercial productions requiring batch generation, JAI Portal's credit system scales efficiently without per-project licensing complications.

Kling O1 Reference Video outputs .mp4 files with resolutions determined by your input video and selected aspect ratio. When using 'auto' aspect ratio, the output matches your input dimensions (typically 720p to 2160p depending on your source). Manual aspect ratios (16:9, 9:16, 1:1) adjust the output resolution accordingly while maintaining quality. The model is optimized for HD and Full HD outputs, making it suitable for web, social media, and broadcast use. All outputs use standard H.264 encoding for broad compatibility. If you need specific resolution control or alternative formats, you can post-process the .mp4 output in your video editor or consider Kling Video v3 Motion Control Pro for additional output customization.

The model analyzes camera motion patterns including pans, tilts, dolly movements, and orbits from your reference video and attempts to replicate that cinematic language in the generated output. Simple, smooth camera movements (slow pan, gentle dolly) are reproduced most accurately. Complex multi-axis movements or rapid cuts may be interpreted more loosely. For best results, use reference videos with intentional, clear camera work rather than accidental handheld shake. If your project requires precise camera path control beyond reference matching, explore Kling Video v3 Motion Control Standart which offers explicit camera trajectory parameters. The reference video approach excels when you want to maintain a specific visual style across multiple shots.

Absolutely. You can submit the same reference video with different prompts, elements, and style images to generate multiple unique variations. This is ideal for A/B testing creative concepts, exploring different narrative directions, or producing a series of shots with consistent motion language but varied content. Each generation is independent, so you can iterate on prompts without affecting previous outputs. The pay-as-you-go credit system makes experimentation affordable—generate several 5-second variations to find the best direction before committing to longer clips. For projects requiring systematic variation testing, consider using JAI Portal's model comparison feature to evaluate Kling Video v2.6 Motion Control Pro alongside reference-based generation for maximum creative flexibility.

⚖️ How Kling O1 Reference Video Compares

Kling O1 Reference Video occupies a unique position among JAI Portal's video generation models by focusing specifically on motion and camera style transfer from reference footage. Unlike text-to-video models that start from scratch, this model excels when you have existing footage whose cinematic language you want to preserve or extend. Compared to Kling Video v3 Motion Control Pro and Kling Video v3 Motion Control Standart, which offer explicit camera path parameters, Kling O1 Reference Video learns motion implicitly from your input, making it faster to set up when you already have a visual reference. For character-focused work, Bytedance Dreamactor v2 provides tighter actor-specific control, but Kling O1 offers more flexibility with scene composition and multiple element integration. Choose this model when you're extending existing footage, maintaining visual continuity across shots, or want to introduce new subjects while preserving a specific cinematic style. It's particularly valuable for filmmakers in pre-visualization, content creators building cohesive video series, and marketers matching brand video aesthetics. For pure text-to-video generation without reference constraints, models like Seedance 2.0 Text to Video offer more creative freedom. Try Kling O1 Reference Video alongside alternatives using JAI Portal's side-by-side comparison feature, or start generating with a free trial at jaiportal.com/auth/signup.

Kling O1 Reference Video

Input Video

Generated Video

Elements Used

More Video Generation Models