Veo 3.1 is the current version of Google's Veo 3 video model family. APIMart serves it through three Ext models (veo3.1-lite, veo3.1-fast, veo3.1-quality, billed per 8-second video) and two official-channel models (veo3.1-fast-official, veo3.1-quality-official, billed per second), all with one API key.
Service Highlights
Video link valid for 72 hours
Transparent pricing with no hidden fees. Pay only for what you use.
| Variant | Tier | GoldCurrent price | Platinum | Diamond | Official | Your savings |
|---|---|---|---|---|---|---|
veo3.1-fast-ext(veo3.1-fast) | 720P/1080P | 1.4Credits(~$0.14)/gen | 1.33Credits(~$0.133)/gen | 1.26Credits(~$0.126)/gen | 1.75Credits(~$0.175)/gen | 20% |
| 4K | 6.4Credits(~$0.64)/gen | 6.08Credits(~$0.608)/gen | 5.76Credits(~$0.576)/gen | 8Credits(~$0.8)/gen | 20% | |
veo3.1-quality-ext(veo3.1-quality) | 720P/1080P | 10Credits(~$1)/gen | 9.5Credits(~$0.95)/gen | 9Credits(~$0.9)/gen | 12.5Credits(~$1.25)/gen | 20% |
| 4K | 15Credits(~$1.5)/gen | 14.25Credits(~$1.425)/gen | 13.5Credits(~$1.35)/gen | 18.75Credits(~$1.875)/gen | 20% | |
veo3.1-lite-ext(veo3.1-lite) | 720P/1080P | 0.7Credits(~$0.07)/gen | 0.665Credits(~$0.0665)/gen | 0.63Credits(~$0.063)/gen | 0.875Credits(~$0.0875)/gen | 20% |
| 4K | 5.7Credits(~$0.57)/gen | 5.415Credits(~$0.5415)/gen | 5.13Credits(~$0.513)/gen | 7.125Credits(~$0.7125)/gen | 20% | |
veo3.1-fast-official | 720P | 0.64Credits(~$0.064)/sec | 0.608Credits(~$0.0608)/sec | 0.576Credits(~$0.0576)/sec | 0.8Credits(~$0.08)/sec | 20% |
| 720P+audio | 0.8Credits(~$0.08)/sec | 0.76Credits(~$0.076)/sec | 0.72Credits(~$0.072)/sec | 1Credits(~$0.1)/sec | 20% | |
| 1080P | 0.8Credits(~$0.08)/sec | 0.76Credits(~$0.076)/sec | 0.72Credits(~$0.072)/sec | 1Credits(~$0.1)/sec | 20% | |
| 1080P+audio | 0.96Credits(~$0.096)/sec | 0.912Credits(~$0.0912)/sec | 0.864Credits(~$0.0864)/sec | 1.2Credits(~$0.12)/sec | 20% | |
| 4K | 2Credits(~$0.2)/sec | 1.9Credits(~$0.19)/sec | 1.8Credits(~$0.18)/sec | 2.5Credits(~$0.25)/sec | 20% | |
| 4K+audio | 2.4Credits(~$0.24)/sec | 2.28Credits(~$0.228)/sec | 2.16Credits(~$0.216)/sec | 3Credits(~$0.3)/sec | 20% | |
veo3.1-quality-official | 720P | 1.6Credits(~$0.16)/sec | 1.52Credits(~$0.152)/sec | 1.44Credits(~$0.144)/sec | 2Credits(~$0.2)/sec | 20% |
| 1080P | 1.6Credits(~$0.16)/sec | 1.52Credits(~$0.152)/sec | 1.44Credits(~$0.144)/sec | 2Credits(~$0.2)/sec | 20% | |
| 720P+audio | 3.2Credits(~$0.32)/sec | 3.04Credits(~$0.304)/sec | 2.88Credits(~$0.288)/sec | 4Credits(~$0.4)/sec | 20% | |
| 1080P+audio | 3.2Credits(~$0.32)/sec | 3.04Credits(~$0.304)/sec | 2.88Credits(~$0.288)/sec | 4Credits(~$0.4)/sec | 20% | |
| 4K | 3.2Credits(~$0.32)/sec | 3.04Credits(~$0.304)/sec | 2.88Credits(~$0.288)/sec | 4Credits(~$0.4)/sec | 20% | |
| 4K+audio | 4.8Credits(~$0.48)/sec | 4.56Credits(~$0.456)/sec | 4.32Credits(~$0.432)/sec | 6Credits(~$0.6)/sec | 20% |
* Actual costs are subject to final output.
Call Google's Veo 3.1 API on APIMart for text to video and image to video. Use veo3.1-lite for low-cost batches, veo3.1-fast for previews and iteration, and veo3.1-quality for final production; switch to the official channel when you need 4, 6 or 8-second clips, an audio switch or a negative prompt. Output runs from 720p to 4K, and finished clips can be extended to 15 seconds.
50K+
Active Users
99.9%
Uptime
2x
Faster
—
Cost Savings
Call Google Veo 3.1 on APIMart with one POST request. Veo 3.1 API generation is asynchronous: submit a task, then poll the task endpoint to get the video URL.
Save the key you created above in the APIMART_API_KEY environment variable and send it as Authorization: Bearer <key> on every Veo 3.1 API request.
POST /v1/videos/generations with model veo3.1-fast, your prompt, aspect_ratio and resolution. Add image_urls and generation_type frame for first and last frame image to video. The Veo 3.1 API responds with a task_id.
Poll GET /v1/tasks/{task_id} every 5–15 seconds until status is completed, then read the video URL from data.result.videos[0].url[0] and save it within 72 hours. Failed Veo 3.1 tasks are not charged.
See the Veo 3.1 API reference for every parameter, the official channel and the remix endpoint.
curl -X POST https://api.apimart.ai/v1/videos/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "veo3.1-fast",
"prompt": "A cinematic drone shot over a misty mountain lake at sunrise, a lone canoe gliding across the water, birdsong and gentle wind",
"duration": 8,
"aspect_ratio": "16:9",
"resolution": "720p"
}'import os
import time
import requests
API_BASE = "https://api.apimart.ai"
headers = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}"}
# 1. Submit the generation task
payload = {
"model": "veo3.1-fast",
"prompt": "A cinematic drone shot over a misty mountain lake at sunrise, a lone canoe gliding across the water, birdsong and gentle wind",
"duration": 8,
"aspect_ratio": "16:9",
"resolution": "720p"
}
resp = requests.post(f"{API_BASE}/v1/videos/generations", headers=headers, json=payload, timeout=60)
resp.raise_for_status()
task_id = resp.json()["data"][0]["task_id"]
# 2. Poll the task until it finishes
for _ in range(200):
task_resp = requests.get(f"{API_BASE}/v1/tasks/{task_id}", headers=headers, timeout=30)
task_resp.raise_for_status()
task = task_resp.json()["data"]
if task["status"] == "completed":
print(task["result"]["videos"][0]["url"][0])
break
if task["status"] == "failed":
raise RuntimeError(task.get("error", {}).get("message", "generation failed"))
time.sleep(10)
else:
raise TimeoutError(f"task {task_id} did not finish in time; continue GET polling with this task ID")curl -X POST https://api.apimart.ai/v1/videos/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "veo3.1-fast",
"prompt": "The camera slowly pushes in from the first frame to the last frame as morning light spreads across the scene, soft ambient sound",
"duration": 8,
"aspect_ratio": "16:9",
"resolution": "720p",
"generation_type": "frame",
"image_urls": [
"https://example.com/first-frame.jpg",
"https://example.com/last-frame.jpg"
]
}'TASK_ID="task_xxx" # replace with the task_id returned by the POST request
curl "https://api.apimart.ai/v1/tasks/$TASK_ID" \
-H "Authorization: Bearer $APIMART_API_KEY"| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Required | Default: veo3.1-fast | veo3.1-lite, veo3.1-fast or veo3.1-quality (the -ext aliases work the same). Official-channel models use the generation-official endpoint parameters. |
prompt | string | Required | Default: — | Video description, including subject, action, camera, lighting and sound. |
duration | integer | Optional | Default: 8 | Fixed at 8 seconds on Ext models. Extend a finished clip to 15 seconds with the remix endpoint. |
aspect_ratio | string | Optional | Default: — | 16:9 (landscape) or 9:16 (portrait). |
resolution | string | Optional | Default: 720p | 720p, 1080p or 4k. |
image_urls | array | Optional | Default: — | Up to 3 public image URLs (JPEG, PNG or WebP, up to 10MB each). Not supported by veo3.1-lite. |
generation_type | string | Optional | Default: — | frame (first image = start frame, second = end frame) or reference (reference images; not on veo3.1-quality). Defaults by image count. Not supported by veo3.1-lite. |
enable_gif | boolean | Optional | Default: false | Return a GIF. Cannot be combined with 1080p or 4k. |
official_fallback | boolean | Optional | Default: false | true uses the official channel as a fallback. Not supported by veo3.1-lite. |
nsfw_check | boolean | Optional | Default: false | Moderate the prompt and input images with omni-moderation-latest before the task is submitted. |
Guides, comparisons and pricing analysis from the APIMart blog.
What you can build with Veo 3.1 on APIMart, and which model or channel supports each option
How businesses and creators use Veo 3.1 API with veo3.1-fast and veo3.1-quality for professional video production
From signup to your first Veo 3.1 video on APIMart
Sign up on APIMart and create an API key on the API Keys page. The key works for every Veo 3.1 model: start with veo3.1-lite or veo3.1-fast for tests, then move to veo3.1-quality or the official channel for final videos. See this page's live pricing table for current prices and discounts.
Describe subject, setting, action, camera movement, lighting and style. Put dialogue in quotation marks and name sound effects, and add a start and end frame or reference images when the shot has to match existing visuals.
Call POST /v1/videos/generations with your model, prompt, aspect ratio and resolution, then poll the task until it completes. Extend a finished Ext clip to 15 seconds with the remix endpoint if you need a longer shot.
What developers and businesses say about Veo 3.1 veo3.1-fast and veo3.1-quality on APIMart
“Veo 3.1 veo3.1-quality model delivers remarkable audio improvement. Our educational Veo 3.1 videos now have natural-sounding dialogue that's perfectly synced with the visuals. Students engage much more with veo3.1-quality content.”
Dr. Emily Zhang
E-Learning Director
“Veo 3.1 scene extension with veo3.1-fast is a game-changer for our marketing campaigns. We create 2+ minute stories by connecting veo3.1-fast clips seamlessly, maintaining character consistency throughout.”
Michael Roberts
Creative Director
“The Veo 3.1 reference image feature in veo3.1-quality lets us maintain brand consistency across all our video content. Upload style guide images and every veo3.1-quality generation matches our visual identity perfectly.”
Sofia Andersson
Brand Manager
“Veo 3.1 veo3.1-fast is perfect for rapid prototyping. We test dozens of variations quickly with veo3.1-fast, then finalize with veo3.1-quality for client deliverables. The Veo 3.1 workflow is incredibly efficient.”
Lucas Huang
Video Producer
“Veo 3.1 cinematic controls in veo3.1-quality are far superior to competitors. Being able to specify exact camera movements and lighting cues in veo3.1-quality gives us the precision we need for professional productions.”
Aisha Patel
Film Director
“With Veo 3.1 veo3.1-fast, 35% better prompt adherence means fewer iterations. Our complex product demo prompts now work the first time with veo3.1-fast, saving time and reducing Veo 3.1 API costs significantly.”
Thomas Klein
Product Marketing Lead
Models, pricing, image to video, extension and access for the Veo 3.1 API on APIMart
"Veo 3 API" (often written veo3 api) means programmatic access to Google's Veo 3 video model family. Veo 3.1 is its current version, and it is what APIMart serves: veo3.1-lite, veo3.1-fast and veo3.1-quality on the Ext channel, plus veo3.1-fast-official and veo3.1-quality-official on the official channel, all under one API key and one set of docs.
Five model IDs. Ext: veo3.1-lite (low-cost, text to video only), veo3.1-fast (previews and iteration) and veo3.1-quality (final production), each also accepted with an -ext suffix. Official: veo3.1-fast-official and veo3.1-quality-official, which add 4/6/8-second durations, negative_prompt, seed and an audio switch.
Ext models are billed per 8-second video by model and resolution (720p and 1080p share a price, 4K costs more); veo3.1-lite is the cheapest. Official models are billed per second, with a higher rate for video with audio. The table on this page shows live Gold member prices (20% off); Platinum and Diamond members save up to 28%. Failed requests are not charged.
Use Ext (veo3.1-lite, veo3.1-fast, veo3.1-quality) for a flat price per 8-second clip, reference-image mode, GIF output and the 15-second remix. Use the official channel (veo3.1-fast-official, veo3.1-quality-official) when you need 4 or 6-second clips, negative_prompt, seed, generate_audio or person_generation control.
Yes. On veo3.1-fast and veo3.1-quality, send up to 3 images in image_urls: generation_type frame uses the first image as the start frame and the second as the end frame, and reference (veo3.1-fast only) uses up to 3 images as references. Images must be public JPEG, PNG or WebP URLs up to 10MB. On the official channel use first_frame_image and last_frame_image. On APIMart, veo3.1-lite is text to video only.
Yes, up to 15 seconds. After an Ext video task succeeds, call POST /v1/videos/{task_id}/remix with the same model (veo3.1-fast or veo3.1-quality) and a continuation prompt. The remix returns a new task ID; set raw to true to get only the extended part. Longer pieces are built by editing several clips together.
All models support 720p (default), 1080p and 4K in 16:9 or 9:16. Ext clips are always 8 seconds, and Ext can return a GIF at 720p. Official clips can be 4, 6 or 8 seconds (8 by default).
Veo 3.1 can generate synchronized dialogue, sound effects and ambience. Write speech in quotation marks ('A man says, "Hello"') and describe sounds explicitly ('SFX: car engine starts'). On the official channel set generate_audio to true to add an audio track, which is billed at the with-audio rate; Ext models have no audio switch.
Sign up for an APIMart account, open the API Keys page and create a key; an API key is required for every Veo request. The same key works as your Veo 3 / Veo 3.1 API key and for every other model on APIMart, with pay-as-you-go billing and no subscription.
POST /v1/videos/generations with a model such as veo3.1-fast and your prompt. Generation is asynchronous: the response contains a task_id, and you poll GET /v1/tasks/{task_id} until the task completes, then read the video from data.result.videos[0].url[0]. Download it within 72 hours. The example section above has copy-ready cURL and Python code.
No. There is no free quota for the Veo 3.1 API on APIMart. Billing is pay-as-you-go: failed requests are not charged, and every price on this page already includes the 20% membership discount.
Google released Veo 3.1 and Veo 3.1 Fast on October 15, 2025, as a paid preview in the Gemini API, Google AI Studio and Vertex AI, and added Veo 3.1 Lite, its most cost-effective Veo model, on March 31, 2026. On APIMart, Veo 3.1 is available through veo3.1-lite, veo3.1-fast, veo3.1-quality and the official channel.
Explore more models in the same category.

Pixverse V6 Max
PixVerse V6 Max is an advanced AI video generation model that transforms text and images into cinematic videos up to 15 seconds long at 1080p. With realistic visuals, expressive characters, and optional synchronized audio, it offers precise creative control for multi-shot storytelling, advertising, and music videos.

Kling V4
Kling 4.0 is an advanced AI video generation model that brings creative ideas to life with realistic motion, dynamic camera work, and precise visual control. With support for continuous videos up to 30 seconds and multiple reference inputs, it helps creators craft cinematic stories with greater flexibility.

ViduQ 4 Preview
Vidu Q4 Preview is ShengShu's image-to-video and reference-to-video model. Animate one first frame or combine up to 15 reference images and 3 audio clips into 3–16 second videos at up to 4K, with audio generated by default.

Wan 3.0 Video Prime
Wan 3.0 Prime is the Prime version of Alibaba Cloud's all-in-one Wan 3.0 video model, with general, first-and-last-frame and file or webpage modes and 2–30 second output.