Premium AI video generation. State-of-the-art motion coherence and cinematic quality.
"Slow-motion close-up of a hummingbird hovering near tropical flowers, golden hour backlight, shallow depth of field, cinematic"
| Model ID |
runway-gen3
|
| Provider | Runway |
| Modality | Video Generation (text-to-video, image-to-video) |
| Max Resolution | 1920×1080 (1080p) |
| Output Duration | 5–10 seconds per generation |
| Average Speed | ~90 seconds |
| Cost | From $0.50 per video |
| Output Format | MP4 |
| Supported Inputs | Text prompt, image + text prompt (img2vid) |
| Key Parameters | prompt, duration, resolution, aspect_ratio, seed, image_url |
| Aspect Ratios | 16:9, 9:16, 1:1, 4:3, 3:4 |
| API Compatibility | OpenAI-compatible endpoint |
| Generation Mode | Asynchronous (webhook or polling) |
Pricing is per-generation. Actual cost may vary by resolution and duration. See full pricing →
Generate video with 8 lines of code. OpenAI-compatible.
from openai import OpenAI
client = OpenAI(
base_url="https://api.fluxpool.ai/v1",
api_key="fp_your_api_key"
)
response = client.videos.generate(
model="runway-gen3",
prompt="A cinematic tracking shot through a neon-lit alley, rain, reflections",
duration=5,
resolution="1080p"
)
# Video generation is async — poll for result
print(response.id) # generation ID
print(response.status) # "processing" → "completed"
print(response.video_url) # available when completed
Video generation is asynchronous. Poll the status endpoint or use webhooks for completion callbacks. Read the docs →
"A cinematic tracking shot through a neon-lit alley, rain, reflections"
Cinematic quality. Commercial-grade motion. Premium output where quality justifies cost.
Motion coherence, camera movement, lighting consistency, human motion, atmospheric effects.
wan-video-2.1 for faster/cheaper video. kling-v2 for character consistency. hunyuan-video for budget video.
Start generating with runway-gen3.