Silent image to video
Elo · silent image-to-video arena
Ray 3.2 overview
Ray 3.2 by Luma AI
Ray 3.2 API delivers production-ready video generation at resolutions from 360p to 1080p across SDR, HDR, and HDR+EXR modes, with editing and reframe capabilities in a single unified model.
Ray 3.2 specs
- Model ID
luma/ray-3.2- Developer
- Luma AI
- Input
- Text, Image, Video
- Output
- Video
- Endpoint
POST https://api.aimlapi.com/v2/video/generations
Ray 3.2 is Luma AI's most advanced video generation model. It introduces HDR and HDR+EXR color modes, multi-resolution output from 360p to 1080p, video editing, and reframe, all through a single API. Compared to Ray 2, Ray 3.2 delivers superior motion coherence, lighting accuracy, and cinematic-grade color grading.
- Supported tasks: Text-to-video, image-to-video, video editing, video reframe
- Resolution: 360p, 540p, 720p, 1080p
- Duration: 5s or 10s (SDR); 5s only (HDR / HDR+EXR)
- Color modes: SDR, HDR, HDR+EXR
- Motion quality: Enhanced camera movement with advanced relighting and temporal consistency
- vs Ray 2: Ray 3.2 adds HDR modes, multi-resolution tiers, video editing, and reframe. Choose Ray 3.2 for professional and cinematic workflows.
- vs Ray Flash 2: Ray Flash 2 is faster and cheaper for high-volume generation. Choose Ray 3.2 when quality is the priority.
Use cases
Cinematic key art and trailers are documented as a use case, including generating hero shots, testing alternate looks on the same blocking, and reframing for theatrical or social formats.
Game cinematics and capsule art are documented as a use case, including animating a hero still for in-game cutscenes or store capsule loops.
Restyling approved footage is documented as a use case, especially when the shot is already approved but needs a different palette or era while preserving the performance.
Aspect ratio variants for cross-platform delivery are documented as a use case, including cutting one master and reframing it for Reels, TikTok, feeds, and premium hero placements.
Ray 3.2 benchmarks by 
Artificial Analysis
Human-preference results for Ray 3.2 in silent image to video, next to models available on AI/ML API. Data retrieved Nov 3, 2026.
Models on AI/ML API
Arena results for models you can call with the same API key.
Silent image to video
Clips animated from a start image, judged as silent video. Ray 3.2 ranks #33 of 34.
Ray 3.2 API pricing
| Usage | Price | Unit |
|---|---|---|
| Video generation | $0.06 | per 1M pixels |
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Ray 3.2 vs other video models: prices
AI/ML API rates for 3 models, per 1M pixels.
| Model | Per 1M pixels |
|---|---|
 Ray 3.2 | $0.06 |
| Ray Flash 2 | $0.0274 |
| Ray 2 | $0.0796 |
How to use the Ray 3.2 API
View docs (opens in a new tab)
Build with your agent
Give your agent the model details, code and documentation.
Paste it into your coding agent to build the integration.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v2/video/generations' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "luma/ray-3.2",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "luma/ray-3.2",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}
response = requests.post("https://api.aimlapi.com/v2/video/generations", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v2/video/generations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "luma/ray-3.2",
"prompt": "A slow dolly shot through a sunlit greenhouse full of ferns, light dust in the air, natural camera motion"
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());The request returns a generation id. Poll GET https://api.aimlapi.com/v2/video/generations?generation_id=<id> until the status is completed.
Models similar to Ray 3.2
Related articles
Ray 3.2 FAQ
What inputs and outputs does Ray 3.2 support?
Ray 3.2 takes image, text, video as input and returns video.
How much does Ray 3.2 cost?
Ray 3.2 is priced at $0.06 / 1M px.
How is Ray 3.2 billed?
Ray 3.2 is billed per generation, a fixed charge per output rather than by prompt length.
Can Ray 3.2 take an existing image as input?
Yes, Ray 3.2 accepts image input as well as a text prompt.
Who created Ray 3.2?
Ray 3.2 was built by Luma AI.
How do I call Ray 3.2 on AI/ML API?
Use luma/ray-3.2 as the model id on AI/ML API.
Is Ray 3.2 available through a single API key?
Yes. Ray 3.2 is served through AI/ML API, so the same key and endpoint format used for other models applies.
What is the Ray 3.2 model ID?
Use luma/ray-3.2 as the model parameter. Requests go to https://api.aimlapi.com/v2/video/generations with your AI/ML API key.