Real outputs from the public examples of this model on Replicate, with the prompt and settings that produced them. They were not generated live on this page.
A hyper-speed superhero, resembling The Flash, is sprinting through a dense, dark forest at night. The trees blur into streaks of green and black as he moves. Fiery trails burst behind him with every stride, igniting parts of the underbrush in glowing embers. As he weaves between the trees, the blazing trail he leaves behind slowly forms the words 'VEO 3 FAST' in glowing, molten fire on the forest floor. The camera zooms up to show the entire text.
Google DeepMind is the merged AI research organisation formed in April 2023 from Google Brain and DeepMind under Demis Hassabis. The Veo video-generation programme delivered Veo 1 (May 2024), Veo 2 (December 2024) and Veo 3 (May 2025), the first major commercial video model with natively generated synchronized audio. Veo 3 Fast was launched alongside Veo 3 as the speed and cost tier: a distilled, low-step variant designed for high-volume creative iteration on Vertex AI, the Gemini API and consumer surfaces such as VideoFX and YouTube Dream Screen. It became the most widely-used Veo variant for high-throughput workflows because the cost difference versus full Veo 3 is significant while quality remains broadly cinematic.
Distilled latent video diffusion / DiT speed tier with joint audio-video diffusion
Veo 3 Fast is the distilled speed-tier variant of Veo 3. Like Veo 3 it is a latent video diffusion model with a transformer-based denoiser plus a coupled audio diffusion module that produces synchronized music, ambient sound and dialogue with lip-sync. Fast is produced via consistency-style sampler distillation and quantization, compressing the diffusion sampler into far fewer denoising steps while preserving most of Veo 3's prompt adherence, motion physics and audio-video sync. Conditioning uses Gemini-family text encoders and optional image embeddings. Native generation produces up to 8-second clips at 1080p with synchronized audio, at noticeably lower latency and per-clip cost than full Veo 3. Training is partly a distillation of Veo 3 over the same curated multilingual audio-video corpus, with auxiliary supervision tuned for sampler-step efficiency.
Speed-tier Veo 3 with native synchronized audio retained
Up to 8-second 1080p clips at substantially lower cost
Text-to-video and image-to-video
Rich cinematographic prompt vocabulary
Multilingual prompts via Gemini text encoders
Available via Vertex AI, Gemini API, VideoFX, Whisk and YouTube Dream Screen
SynthID audio + visual watermarking
Materially lower latency than full Veo 3
Best for: high-volume creative iteration, ad A/B testing, social shorts with sound.
Entraînement et licence
Distilled from Veo 3 over the same curated multilingual audio-video corpus, with auxiliary supervision tuned for sampler-step efficiency and audio-video synchronization.
Licence: Proprietary commercial licence via Google Cloud / Vertex AI and Gemini API; commercial use under Google's generative-AI terms; mandatory SynthID watermarking.
curl https://railwail.com/api/v1/videos/generations \
-H "Authorization: Bearer $RAILWAIL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "veo-3-fast",
"prompt": "A slow drone shot over a misty pine forest at sunrise"
}'
# Response: {"job_id": "...", "status": "queued", ...}
# Poll until status is completed, failed or cancelled:
curl https://railwail.com/api/v1/jobs/JOB_ID \
-H "Authorization: Bearer $RAILWAIL_API_KEY"
import os
import time
import requests
API = "https://railwail.com/api/v1"
headers = {"Authorization": f"Bearer {os.environ['RAILWAIL_API_KEY']}"}
job = requests.post(
f"{API}/videos/generations",
headers=headers,
json={
"model": "veo-3-fast",
"prompt": "A slow drone shot over a misty pine forest at sunrise",
},
).json()
while True:
status = requests.get(f"{API}/jobs/{job['job_id']}", headers=headers).json()
if status["status"] in ("completed", "failed", "cancelled"):
break
time.sleep(5)
print(status["status"], status.get("output_url"))
const API = "https://railwail.com/api/v1";
const headers = {
Authorization: `Bearer ${process.env.RAILWAIL_API_KEY}`,
"Content-Type": "application/json",
};
const job = await fetch(`${API}/videos/generations`, {
method: "POST",
headers,
body: JSON.stringify({
model: "veo-3-fast",
prompt: "A slow drone shot over a misty pine forest at sunrise"
}),
}).then((r) => r.json());
let status;
do {
await new Promise((r) => setTimeout(r, 5000));
status = await fetch(`${API}/jobs/${job.job_id}`, { headers }).then((r) => r.json());
} while (!["completed", "failed", "cancelled"].includes(status.status));
console.log(status.status, status.output_url);
Proprietary commercial licence via Google Cloud / Vertex AI and Gemini API; commercial use under Google's generative-AI terms; mandatory SynthID watermarking.
