Vidu Q2 · Image Generation / 参照から画像
Next-gen image-to-video model with "emotional acting" for nuanced facial expressions. Supports 2–8s clips at 1080p with cinematic camera control. Features multi-modal reference (2 videos + 4 images) enabling precise replication of expressions, actions, textures, and effects. Includes video editing: element add/delete/replace, style transfer, aspect adjustment. Lightning mode outputs 5s in 20s. Supports up to 7 consistent subjects, 3x faster than Q1
Vidu Q2Reference-to-Image at 1080p–4K
Vidu Q2 generates stills from text or references at 1080p, 2K, or 4K with nine aspect ratios and optional seed control.
At a Glance
Vidu Q2 Capabilities
Reference-aware stills with multi-tier fidelity.
Reference Conditioning
Feed reference images so identity and style stay locked across generated stills.
1080p / 2K / 4K Tiers
Draft at 1080p, ship at 2K, and promote print or OOH heroes to 4K.
Nine Aspect Ratios
16:9, 9:16, 1:1, 3:4, 4:3, 21:9, 2:3, 3:2, and auto cover social, print, and cinematic placements.
Seed Reproducibility
Pin a seed to regenerate the same creative baseline after brief revisions.
How It Works
From brief to reference-aware still.
Write the Brief
Describe subject, style, and use-case.
Optional References
Add reference images to lock identity and style.
Pick Fidelity & Ratio
1080p/2K/4K; choose from nine ratios or auto.
Pin Seed if Needed
Lock a seed for reproducible campaign frames.
Vidu Q2 Domains
Where reference stills and multi-tier fidelity matter.
Character Campaigns
Lock talent identity across still variants.
Social Carousels
Multi-ratio frames with consistent style.
Print Masters
4K output for large-format placements.
Mood Boards
Reference-conditioned exploration boards.
Prompt Tips
Better Vidu Q2 stills on the first pass.
Reference images work best when each has a single clear job.
Draft at 1080p and promote final masters to 4K.
Reuse a seed so small brief edits do not re-roll the entire composition.
Vidu Q2 Quickstart
Reference-aware still generation.
curl -X POST "https://api.powertokens.ai/v1/images/generations" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type": "application/json" \
-d '{
"model": "viduq2",
"prompt": "Editorial close-up portrait with sculptural hat, photorealistic studio light.",
"size": "2048",
"response_format": "url",
"watermark": false,
"aspect_ratio": "16:9"
}'Technical Specifications
Confirmed parameters and runtime execution protocols.
料金詳細
このモデルの実際の課金は、API リクエストで渡される特定のパラメータに基づいて動的に計算されます。以下は具体的な組み合わせとそれに対応する料金です：
| モダリティ | クレジット | 料金 (USD) |
|---|---|---|
| Text to Image1080P | 29/ Image | $0.029 |
| Text to Image2K | 38/ Image | $0.038 |
| Text to Image4K | 48/ Image | $0.048 |
| Reference to Image1080P1-3 Images | 38/ Image | $0.038 |
| Reference to Image1080P4-7 Images | 48/ Image | $0.048 |
| Reference to Image2K1-3 Images | 57/ Image | $0.057 |
| Reference to Image2K4-7 Images | 76/ Image | $0.076 |
| Reference to Image4K1-3 Images | 95/ Image | $0.095 |
| Reference to Image4K4-7 Images | 143/ Image | $0.143 |
Recommended Related Models
Explore complementary video and multimodal models with your unified API key.
wan2.7-image
wan2.7-image
Alibaba's unified image generation and editing model supporting text-to-image, sequential image generation, instruction-based editing, multi-reference generation (up to 9 images), and interactive bbox editing. Features precise "face sculpting" for diverse, non-generic portraits and a "palette" function for exact Hex color control. Renders up to 3K tokens of long-form text (12 languages) at print quality. Generates up to 12 images per run
Seedream 5.0
seedream-5-0-260128
Supports text, single-image and multi-image inputs, and enables the generation of image sets
Wan 2.7 Image Pro
wan2.7-image-pro
Wan2.7–image-pro，supports text to image, text/image to sequential images, image editing, multi-image reference generation, and interactive editing. Delivers enhanced performance in text rendering, subject consistency, and complex instruction following.
Kling Image O1
kling-image-o1
Advanced image generation model from the Kling O1 family. Supports text-to-image and image-to-image workflows with deep semantic understanding. Features character/subject consistency across generations, multi-reference input (up to 7 images), and intelligent aspect ratio adaptation. Professional-grade output suitable for commercial creative pipelines
Frequently Asked Questions
Everything you need to know before integrating this model.
Start Building with Vidu Q2 Today
Create an account in seconds to receive free test credits. No upfront payment or long-term contract required.