Why teams evaluate this API
Built for developers who want broad Vidu API coverage, exact preflight pricing, one API key, and a consistent asynchronous task lifecycle without building a separate media-processing layer.
Video, Image & Audio
Vidu API relay with all 27 endpoints and complete pricing
Access 27 Vidu image-to-video, text-to-video, image, audio, motion, template, and film APIs through one URL-only task contract with complete pricing.
Best-fit teams
- • AI video and creator products that need direct Vidu model access
- • Automation platforms combining video, image, audio, and speech workflows
- • Teams that want quote-first spend controls and one task contract across Vidu features
Core capabilities
- • Q3, Q2, Q1, and Vidu 2.0 image-, text-, reference-, and start/end-to-video workflows
- • Reference and text-to-image, text/timing-to-audio, TTS, and voice cloning
- • Lip-sync, motion transfer, extension, multi-frame, upscale, replacement, and digital humans
- • Templates, prompt recommendations, general films, advertising films, and trending replication
- • Fixed operation allowlist, URL-only media inputs, quote-before-create, polling, cancellation, and webhooks
Typical implementation flow
- 1Select one allowlisted operation and put its native Vidu fields inside input
- 2Call POST /api/v1/vidu/quote to calculate the current model and option price
- 3Create the task with an Idempotency-Key and optional maxCredits guard
- 4Poll the task or receive a signed webhook, then copy the temporary result URL promptly
Complete public relay catalog
27 Vidu operations, inputs, endpoints, and prices
Every Direct task is quoted from the selected operation, model, resolution, duration, and options. One Vidu provider credit equals 2 OmnAPI credits. The effective USD amount depends on your credit package. The Music Video product has a separate pricing scope; no MV-specific customer rate is applied to the operations below.
- Quote
- POST /api/v1/vidu/quote
- Create
- POST /api/v1/vidu/tasks
- Status
- GET /api/v1/vidu/tasks/{id}
- Cancel
- POST /api/v1/vidu/tasks/{id}/cancel
Complete operation directory
Endpoint paths are fixed Vidu Enterprise v2 routes selected by OmnAPI. Put Vidu request fields inside the input object; callers cannot submit an arbitrary upstream path or method.
| Category | Operation | Vidu path | What it does | Required input | OmnAPI pricing |
|---|---|---|---|---|---|
| Video | image-to-video | /img2video | Animate one source image with Q3, Q2, Q1, or Vidu 2.0. | model, images[1] | Video model matrix; optional audio +15 Vidu credits on Q2 |
| Video | reference-to-video | /reference2video | Generate a subject- or scene-consistent video from references. | model, subjects[], prompt | Reference-video matrix; optional audio +15 Vidu credits on Q2 |
| Video | start-end-to-video | /start-end2video | Generate motion between a start frame and an end frame. | model, images[2] | Image/start-end video matrix |
| Video | text-to-video | /text2video | Generate video directly from a text prompt. | model, prompt | Text-video model matrix |
| Image | reference-to-image | /reference2image | Run text-to-image, reference generation, or image editing. | model, prompt | By model, output resolution, and reference-image count |
| Audio | text-to-audio | /text2audio | Generate a sound effect or short audio clip from text. | model, prompt | 20 OmnAPI credits up to 5s; 40 up to 10s |
| Audio | timing-to-audio | /timing2audio | Generate multiple timed or overlapping sound events. | model, timing_prompts[] | 20 OmnAPI credits up to 5s; 40 up to 10s |
| Speech | text-to-speech | /audio-tts | Synthesize speech with a Vidu voice and voice controls. | text, voice_setting_voice_id | 20 OmnAPI credits per started 500 characters; minimum 20 |
| Speech | voice-clone | /audio-clone | Clone a voice and synthesize its required preview. | audio_url, voice_id, text | 600 OmnAPI credits per clone + TTS preview charge |
| Video edit | lip-sync | /lip-sync | Synchronize a speaking video to supplied text or audio. | video_url + audio_url or text | 40 OmnAPI credits per started 5 seconds |
| Motion | motion-sync-v1 | /image2dance | Transfer reference-video motion to a source image. | image_url, video_url | 20 OmnAPI credits per started second |
| Motion | motion-sync-v2 | /template | Run current Vidu motion-control templates. | images, video_urls | 38 credits/s for motion_control_2; 68 credits/s for 2.5 |
| Video edit | extend-video | /extend | Extend an existing Vidu creation or public video by 1–7 seconds. | model + video_creation_id or video_url | Q2 extension matrix |
| Video edit | multi-frame-video | /multiframe | Generate a consistent video across multiple keyframes. | model, start_image, image_settings[] | Sum of the Q2 price for every frame-to-frame segment |
| Video edit | upscale-video | /upscale-new | Upscale a video to 1080p, 2K, 4K, or 8K. | video_url or video_creation_id | 20 / 40 / 80 / 320 OmnAPI credits per second |
| Video edit | replace-video | /replace | Replace a target object, character, or region in a video. | video_url, object, image | Authorization cap; settle reported actual usage at 2× |
| Character | digital-human | /digital-human | Generate a speaking digital-human video from an image. | model, image + selected speech-mode fields | Authorization cap; settle reported actual usage at 2× |
| Prompt | prompt-recommendation | /img2video-prompt-recommendation | Recommend prompts for image-to-video or template use. | images, type[] | 20 OmnAPI credits per started batch of 5 prompts per type |
| Template | template | /template | Apply a Vidu effect template to supplied images. | template, images[] | Authorization cap because price varies by template |
| Template | template-story | /template-story | Generate a longer coherent video from a story template. | story, images[] | Authorization cap because price varies by story template |
| Film | general-film-create | /one-click/general_one_click | Create a general-purpose one-click finished film. | images[], duration | 80 OmnAPI credits per second |
| Film | general-film-edit | /one-click/general_one_click/{id}/edit | Edit footage, narration, or BGM in a general film. | id, job_id, prompt | 80/s footage; 20/500 chars narration; 4/s BGM |
| Film | general-film-compose | /one-click/general_one_click/{id}/compose | Compose selected general-film edit jobs. | id, job_ids | 2 OmnAPI credits per call |
| Advertising | ad-film-create | /ad-one-click | Create a one-click advertising film. | images[] | 80 OmnAPI credits per second; default duration 15s |
| Advertising | ad-film-edit | /ad-one-click/edit | Edit advertising footage, narration, or BGM. | ad_one_click_task_id, type, prompt | 80/s footage; 20/500 chars narration; 4/s BGM |
| Advertising | ad-film-compose | /ad-one-click/compose | Compose selected advertising-film jobs. | ad task ID, video task IDs, BGM and narration IDs | 2 OmnAPI credits per call |
| Trends | trending-replicate | /trending-replicate | Recreate the format and motion of a trending source video. | video_url, images[] | 24/s at 540p; 32/s at 720p; 40/s at 1080p |
Complete price matrix
Vidu credits explain the upstream rate-card quantity. OmnAPI credits are the amount authorized from your balance. The quote endpoint is the final source for an individual request and applies whole-credit rounding.
Q3 image, text, and start/end video
Price is per generated second. Normal / off-peak values are shown together.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q3 Pro | 540p | 9 / 5 per second | 18 / 10 per second |
| Q3 Pro | 720p | 20 / 10 per second | 40 / 20 per second |
| Q3 Pro | 1080p | 24 / 12 per second | 48 / 24 per second |
| Q3 Turbo | 540p | 7 / 4 per second | 14 / 8 per second |
| Q3 Turbo | 720p | 11 / 6 per second | 22 / 12 per second |
| Q3 Turbo | 1080p | 13 / 7 per second | 26 / 14 per second |
| Q3 Pro Fast | Image-to-video · 720p | 20 / 10 per second | 40 / 20 per second |
| Q3 Pro Fast | Image-to-video · 1080p | 25 / 13 per second | 50 / 26 per second |
Q3 reference-to-video
Price is per generated second. Q3 Mix does not support off-peak pricing.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q3 Mix | 720p | 24 per second | 48 per second |
| Q3 Mix | 1080p | 29 per second | 58 per second |
| Q3 Turbo | 540p | 4 / 2 per second | 8 / 4 per second |
| Q3 Turbo | 720p | 10 / 5 per second | 20 / 10 per second |
| Q3 Turbo | 1080p | 13 / 7 per second | 26 / 14 per second |
| Q3 | 540p | 7 / 4 per second | 14 / 8 per second |
| Q3 | 720p | 12 / 6 per second | 24 / 12 per second |
| Q3 | 1080p | 15 / 7 per second | 30 / 14 per second |
Q2 video generation
s is the requested duration rounded up to a whole second. Off-peak is half the Vidu total, rounded up, then converted at 2×. Q2 image/reference audio adds 15 Vidu or 30 OmnAPI credits.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q2 Turbo image/start-end | 540p | 6 + 2×(s−1) | 12 + 4×(s−1) |
| Q2 Turbo image/start-end | 720p | s=1: 8; s≥2: 10 + 10×(s−2) | s=1: 16; s≥2: 20 + 20×(s−2) |
| Q2 Turbo image/start-end | 1080p | 35 + 10×(s−1) | 70 + 20×(s−1) |
| Q2 Pro image/start-end | 540p | s=1: 8; s≥2: 10 + 5×(s−2) | s=1: 16; s≥2: 20 + 10×(s−2) |
| Q2 Pro image/start-end | 720p | 15 + 10×(s−1) | 30 + 20×(s−1) |
| Q2 Pro image/start-end | 1080p | 55 + 15×(s−1) | 110 + 30×(s−1) |
| Q2 Pro Fast image/start-end | 720p | 8 + 2×(s−1) | 16 + 4×(s−1) |
| Q2 Pro Fast image/start-end | 1080p | 16 + 4×(s−1) | 32 + 8×(s−1) |
| Q2 text-to-video | 540p | 10 + 2×(s−1) | 20 + 4×(s−1) |
| Q2 text-to-video | 720p | 15 + 5×(s−1) | 30 + 10×(s−1) |
| Q2 text-to-video | 1080p | 20 + 10×(s−1) | 40 + 20×(s−1) |
| Q2 reference-to-video | 540p / 720p / 1080p | 15/25/75 + 5/5/10×(s−1) | 30/50/150 + 10/10/20×(s−1) |
| Q2 Pro reference-to-video | 540p / 720p / 1080p | 20/30/85 + 5/5/10×(s−1) | 40/60/170 + 10/10/20×(s−1) |
Q1 and Vidu 2.0 video
Fixed supported duration and resolution combinations; normal / off-peak.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q1 | 5s · 1080p · image/reference/start-end/text | 80 / 40 | 160 / 80 |
| Vidu 2.0 image/start-end | 4s · 360p | 20 / 10 | 40 / 20 |
| Vidu 2.0 image/start-end | 4s · 720p | 40 / 20 | 80 / 40 |
| Vidu 2.0 image/start-end | 4s · 1080p | 100 / 50 | 200 / 100 |
| Vidu 2.0 image/start-end | 8s · 720p | 100 / 50 | 200 / 100 |
| Vidu 2.0 reference | 4s · 360p or 720p | 80 / 40 | 160 / 80 |
Image, audio, and speech
Image price is per output. Speech uses the current-credit equivalent of the dollar price where Vidu's public credit column still shows the old denomination.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q2 text-to-image | 1080p / 2K / 4K | 6 / 8 / 10 | 12 / 16 / 20 |
| Q2 reference image · 1–3 refs | 1080p / 2K / 4K | 8 / 12 / 20 | 16 / 24 / 40 |
| Q2 reference image · 4–7 refs | 1080p / 2K / 4K | 10 / 16 / 30 | 20 / 32 / 60 |
| Q1 reference image · 1–7 refs | 1080p | 20 | 40 |
| Text/timing-to-audio | ≤5s / >5s to 10s | 10 / 20 | 20 / 40 |
| Text-to-speech | Each started 500 characters | 10; minimum 10 | 20; minimum 20 |
| Voice clone | Per voice + required TTS preview | 300 + TTS | 600 + TTS |
Video tools and motion
For URL-priced tools, provide pricingContext.durationSec because OmnAPI does not download or inspect media.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Prompt recommendation | Each started 5 prompts, per requested type | 10 | 20 |
| Lip sync | Each started 5 seconds | 20 | 40 |
| Motion Sync V1 | Each started second · 720p | 10 | 20 |
| Motion Control 2 | Each started second · 720p | 19 | 38 |
| Motion Control 2.5 | Each started second · 1080p | 34 | 68 |
| Upscale Pro | 1080p / 2K / 4K / 8K per second | 10 / 20 / 40 / 160 | 20 / 40 / 80 / 320 |
| Template / Story / Replace / Digital human | Operation-specific | Caller authorization cap | cap × 2; actual usage settled |
Video extension and multi-frame
s is each extension or segment duration rounded up to a whole second. Multi-frame totals every segment independently.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| Q2 Turbo extension | 540p | 10 + 2×(s−1) | 20 + 4×(s−1) |
| Q2 Turbo extension | 720p | 15 + 5×(s−1) | 30 + 10×(s−1) |
| Q2 Turbo extension | 1080p | 40 + 10×(s−1) | 80 + 20×(s−1) |
| Q2 Pro extension | 540p | 15 + 5×(s−1) | 30 + 10×(s−1) |
| Q2 Pro extension | 720p | 30 + 10×(s−1) | 60 + 20×(s−1) |
| Q2 Pro extension | 1080p | 60 + 15×(s−1) | 120 + 30×(s−1) |
| Multi-frame | Up to 9 frame-to-frame segments | Sum Q2 segment prices | 2 × summed Vidu credits |
Film and advertising solutions
The general-film and ad-film APIs share the same create/edit/compose rates.
| Model or feature | Scope | Vidu credits | OmnAPI credits |
|---|---|---|---|
| General film / Ad film create | 1080p · per second | 40 | 80 |
| Edit footage | 1080p · per second | 40 | 80 |
| Edit narration | Each started 500 characters | 10 | 20 |
| Edit BGM | Per second | 2 | 4 |
| Compose film | Per call | 1 | 2 |
| Trending replicate | 540p / 720p / 1080p per second | 12 / 16 / 20 | 24 / 32 / 40 |
URL-only data boundary
OmnAPI relays public HTTP(S) media URLs and does not upload, download, probe, transcode, or store customer media for Vidu Direct. Provide pricingContext.durationSec when URL media duration affects price.
Variable-price operations
Template, template-story, replace-video, and digital-human use pricingContext.providerCreditCap. OmnAPI reserves cap × 2 and refunds unused credits when Vidu reports lower actual usage.
quote → create → poll/webhookFAQ
How much does the Vidu API cost through OmnAPI?
Every Vidu provider credit is billed as 2 OmnAPI credits. The total depends on the operation, model, resolution, duration, off-peak option, and add-ons. The page lists both the Vidu rate-card quantity and the corresponding OmnAPI amount, and the quote endpoint returns the exact current authorization before creation.
Which Vidu APIs are available?
The Direct product exposes 27 operations covering image-, reference-, start/end-, and text-to-video; reference-to-image; audio and speech; lip-sync and motion; video extension, multi-frame, upscale, and replacement; digital humans and prompts; templates; general films, ad films, and trending replication.
Does Vidu Direct use the Music Video API customer price?
No. Vidu Direct is a separate product scope and always uses its own 2 OmnAPI credits per Vidu credit public rate. Music Video pricing and any customer-specific MV agreement do not apply to Direct operations.
Does OmnAPI upload or process Vidu media?
No. Vidu Direct is a URL-only relay. It does not upload, download, inspect, transcode, or store customer media. Supply public HTTP(S) URLs and provide the known duration in pricingContext when a URL-based operation is duration-priced.
How are Vidu templates and other variable-price calls billed?
Template, template-story, replace-video, and digital-human require a provider-credit authorization cap because Vidu does not publish one deterministic rate for every variant. OmnAPI reserves cap × 2 and refunds unused credits when reported actual usage is lower.
Is Vidu S1 real-time streaming included?
No. Vidu S1 requires a live WebSocket and RTC media session, while this product is a stateless asynchronous REST relay. Vidu upload endpoints are also excluded because callers provide public media URLs.