Use Vidu Q3 online, pay per video

What Vidu Q3 does well

Vidu Q3 is Shengshu's video model, offered on Kreavo in two versions. Vidu Q3 Pro is described as premium quality with native audio, up to 16 seconds. Vidu Q3 Turbo is described as faster generation, native audio, same quality, up to 16 seconds. Both accept up to 2 reference images, output 540p, 720p or 1080p, and support five aspect ratios. The 16-second ceiling with native audio is the notable spec: among the audio-capable models on Kreavo, Vidu Q3 runs longer than VEO 3.1 (fixed 8 seconds) and Kling 3.0 (up to 15 seconds).

• Native audio, on by default
• 5, 8, 10, 12 or 16 second clips
• 540p, 720p or 1080p output
• Up to 2 reference images per generation
• Aspect ratios 16:9, 9:16, 4:3, 3:4 and 1:1

Vidu Q3 pricing on Kreavo

Kreavo is pay-as-you-go: 1,000 credits cost $10, so one credit is $0.01 — the top-up rate. Monthly plans can be cheaper. Both Vidu Q3 versions share one credit balance with every other model, and there is no Shengshu plan to buy.

• Vidu Q3 Pro and Vidu Q3 Turbo: from $0.75 per video (540p) up to $5.60 (1080p, longest clip)
• VEO 3.1 Fast (audio, fixed 8s): $2.00 per video
• Kling 3.0 (native audio, up to 15s): from $1.40 to $4.20 per video
• Free trial: 150 credits, no credit card

Vidu Q3 vs other models on Kreavo

Vidu Q3 is the entry point for video with native audio on Kreavo: a 540p clip starts at $0.75, below VEO 3.1 Fast from Google DeepMind at $2.00 and Kling 3.0 from Kuaishou at $1.40. Kling 3.0 answers with multi-shot storytelling and 4K on its Pro tier ($1.80 to $5.60); VEO 3.1 answers with Google's rendering and a 4K option. For silent clips beyond 16 seconds, Seedance 2.5 from ByteDance ($1.75) and Wan 3.0 from Alibaba go to 30 seconds. With every model in one account, the same two reference images can be run through Vidu Q3 Pro, Vidu Q3 Turbo and Kling 3.0 in a few minutes.

Best use cases for Vidu Q3

• AI UGC with sound up to 16 seconds: a spoken hook, product handling noise, ambient audio
• Product videos with your real product as a reference image, composited faithfully
• Consistent AI influencer clips using the second reference slot for the face
• Cheap audio drafts at 540p for $0.75, then a 1080p final on the winning script
• Square 1:1 and 4:3 formats that many video models do not offer

Limits to know

• Clip length caps at 16 seconds; Seedance 2.5 and Wan 3.0 reach 30 seconds without audio
• Maximum resolution is 1080p; Kling 3.0 Pro and VEO 3.1 are the 4K options
• Up to 2 reference images per generation
• Durations are fixed steps (5, 8, 10, 12, 16 seconds), not a free slider