Kling Video O1 Reference-to-Video overview
Kling Video O1 Reference-to-Video by Kling AI
It uses advanced feature extraction to preserve visual identity such as appearance, texture, and style across entirely new scenarios and motions.
Kling Video O1 Reference-to-Video specs
- Model ID
kling-video/o1/reference-to-video- Developer
- Kling AI
- Input
- Text, Image
- Output
- Video
- Endpoint
POST https://api.aimlapi.com/v2/video/generations
Kling Video O1 API Overview
Kling Video O1 Reference-to-Video by Kuaishou delivers breakthrough subject-consistent video generation from image references. This unified multimodal model preserves character, prop, and scene identity across entirely new scenarios, powered by advanced feature extraction.
Technical Specifications
- Input Support: Single or multiple reference images (up to 4 viewpoints per element) in JPG, JPEG, PNG formats; optional video references up to 10s, 200MB, 2K resolution.
- Output Capabilities: Videos from 5-10 seconds; resolutions up to 2K (1080p standard); 30fps; aspect ratios including 16:9.
- Model Architecture: Unified multimodal engine with Chain of Thought (CoT) reasoning, multi-element fusion, and vision-language processing for precise identity retention.
Performance Benchmarks
Kling Video O1 excels in identity consistency and motion quality. Internal tests show 247% improvement over Google Veo 3.1 and 230% over Runway Aleph in reference generation tasks.
- Superior frame stability reduces flickering in multi-subject scenes.
- Enhanced reasoning via CoT boosts prompt accuracy by analyzing inputs before rendering.
Key Features
- Multi-reference subject building extracts features from various viewpoints to ensure stable identity in dynamic scenes.
- New scenario generation creates fresh content like futuristic walks or interactions while locking in reference details.
- Professional/Standard modes balance quality and speed; supports camera control, motion accuracy, and physics simulation.
- All-in-one reference handling fuses multiple subjects (characters, props, scenes) for complex, consistent outputs.
Model Comparisons
vs Google Veo 3.1: Kling O1 outperforms by 247% in reference fidelity, offering better multi-view fusion without coherence loss; Veo lags in complex subject interactions.
vs Runway Gen-4.5: Superior identity retention across angles makes Kling ideal for pro-grade consistency; Runway focuses more on text-driven motion but struggles with multi-references.
vs Hailuo 2.3: Kling's CoT reasoning delivers smoother physics and camera work; Hailuo excels in speed but trails in subject stability for extended clips.
Use cases
- Character Animation: Animate consistent avatars in new environments, ideal for game devs or social media creators.
- Product Videos: Generate demos with fixed objects in varied settings for e-commerce marketing.
- Scene Recreation: Build cinematic sequences from reference photos, preserving props and backgrounds for filmmakers.
Kling Video O1 Reference-to-Video API pricing
| Usage | Price | Unit |
|---|---|---|
| Video generation | $0.112 | per second |
The final price depends on request parameters such as resolution or duration.
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Kling Video O1 Reference-to-Video vs other video models: prices
AI/ML API rates for 5 models, per second.
| Model | Per second |
|---|---|
 Kling Video O1 Reference-to-Video | $0.112 |
| Kling 4.0 Image to Video | $0.157 |
| Kling 4.0 Reference to Video | $0.157 |
| Kling 4.0 Text to Video | $0.157 |
| Kling 4.0 Video Editing | $0.188 |
How to use the Kling Video O1 Reference-to-Video API
View docs
Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v2/video/generations' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "kling-video/o1/reference-to-video",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "kling-video/o1/reference-to-video",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}
response = requests.post("https://api.aimlapi.com/v2/video/generations", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v2/video/generations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "kling-video/o1/reference-to-video",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());The request returns a generation id. Poll GET https://api.aimlapi.com/v2/video/generations?generation_id=<id> until the status is completed.
Models similar to Kling Video O1 Reference-to-Video
Related articles
Kling Video O1 Reference-to-Video FAQ
What inputs and outputs does Kling Video O1 Reference-to-Video support?
Kling Video O1 Reference-to-Video takes image as input and returns video.
How much does Kling Video O1 Reference-to-Video cost?
Kling Video O1 Reference-to-Video costs $0.112 per second on AI/ML API, billed as you go with no subscription.
How is Kling Video O1 Reference-to-Video billed?
Kling Video O1 Reference-to-Video costs $0.112 per second on AI/ML API, billed as you go with no subscription.
Can Kling Video O1 Reference-to-Video take an existing image as input?
Yes, Kling Video O1 Reference-to-Video accepts image input as well as a text prompt.
Who created Kling Video O1 Reference-to-Video?
Kling Video O1 Reference-to-Video was built by Kling AI.
How do I call Kling Video O1 Reference-to-Video on AI/ML API?
Send a request to v2/video/generations with kling-video/o1/reference-to-video as the model id.
Is Kling Video O1 Reference-to-Video available through a single API key?
Yes. Kling Video O1 Reference-to-Video is served through AI/ML API, so the same key and endpoint format used for other models applies.
How does Kling Video O1 Reference-to-Video keep the subject's appearance consistent across a generated clip?
It uses AI reasoning to maintain consistent appearance based on the reference images provided as input.
What is the Kling Video O1 Reference-to-Video model ID?
Use kling-video/o1/reference-to-video as the model parameter. Requests go to https://api.aimlapi.com/v2/video/generations with your AI/ML API key.