Skip to content
← Models
ClosedShengshu Technology/10 trustMetrics-checked

Vidu Q1

Reference-guided video with first-to-last-frame transitions and text-generated audio — 400M+ videos served.

Visit official page →

About

Vidu Q1 (April 21, 2025) is Shengshu Technology's reference-guided video model: first-to-last-frame transitions (upload two images, get a cinematic transition), up to 7 reference images for multi-subject consistency, and text-generated 48kHz background music/foley inside the same workflow — plus 5s 1080p output and improved anime generation.

Closed and commercial: web app + developer platform, credit-based. Scale is real: 400M+ videos generated, 200+ countries, $20M+ ARR within eight months of launch (per company reports, Sept 2025). VBench-internal scores claimed ahead of Runway Gen-2 and Sora on prompt fidelity — treat company benchmarks skeptically. Tsinghua-backed; a serious player in the reference-guided niche alongside Kling.

Specs

Context window

N/A (text/image-to-video; up to 7 reference images)

Pricing

Commercial credit-based pricing (see vidu.studio); developer platform available.

Free quota

Free trial credits; verify current allowance.

API access

Vidu web app (vidu.studio) and developer platform.

Weights

Closed — API only

Released

April 21, 2025

Benchmarks

Company-claimed (VBench internal): ahead of Runway Gen-2 and Sora on fidelity/coherence — treat as company-claimed. Independent standardized benchmarks are limited.

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

videotext-to-videoimage-to-videofilmmakingmarketing

Trust score — 7/10

/10 trustMetrics-checked

Scored 7/10 (metrics tier). Reference-guided generation with first-to-last-frame transitions is a real differentiator, and the scale figures (400M videos, $20M ARR) indicate genuine production usage. Docked three points: (1) closed and commercial; (2) benchmark claims are company-reported VBench runs; (3) the reference-guided niche overlaps heavily with Kling — evaluate both for consistency needs.

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
8.0/10
Maintenance · recency, releases, issue load
5.0/10
Openness · license + docs
4.2/10
Manual review · sanity read + red-flag check
8.0/10
  • · provider-tier estimate — thinner evidence (no public adoption metric)
  • · 400M+ videos generated, $20M+ ARR (company-reported); no public adoption metric
  • · released 540 days ago
  • · manual review: first-to-last-frame transitions are a real differentiator; overlaps Kling's reference-guided niche; license: proprietary — web app + developer platform; credit-metered

Sources