ai/ltx-2.3-spicy/image-to-video-lora
lightricks/ltx-2.3-spicy/image-to-video-lora
LTX 2.3 Spicy LoRA Image to Video API turns a reference image and prompt into expressive AI videos using selectable LoRA presets, optional LoRA strength overrides, duration, and resolution controls. Run fast REST inference on MaaS with no cold starts and affordable pricing.
Examples
Parameters
| Name | Type | Default | Constraints | Description |
|---|---|---|---|---|
| image *Image | image_upload | — | image/* · 1–1 items | Reference image for the first frame (JPG/PNG recommended). |
| promptPrompt | textarea | — | ≤ 5000 chars | Positive prompt describing the desired motion / scene. |
| presetPreset | select | tuned | tuned | original | LoRA preset bundle. `original` = light styling. `tuned` = the recommended general-purpose stack. Overridden per-LoRA via `loras`. |
| lorasLoras | array<object> | — | 0–3 items | List of LoRAs to apply (max 3). |
| ↳path *Path | select | — | sulphur | sulphur_v1 | vbvr | dreamly | synth | plora | singularity | omninft | … | LoRA short key. |
| ↳scaleScale | number | 1 | — | Strength applied to the LoRA. |
| resolutionResolution | select | 480p | 480p | 720p | 1080p | Output resolution tier. |
| durationDuration | slider | 5 | 3 ~ 20 · step 1 | Output duration in seconds (3-20). |
| seedSeed | number | — | — | Random seed. -1 for a random seed. |
Output fields
| Field | Type | Description |
|---|---|---|
| videos | array<string> | Generated video URL(s). |
API
Call this model through one unified REST API. Get a key on the API Keys page.
cURL
# 1) Submit — returns { "task_uuid": "..." }
curl -X POST "https://www.namifusion.com/api/v1/marketplace/run/lightricks/ltx-2.3-spicy/image-to-video-lora" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"image": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
],
"prompt": "Bring the reference image to life with subtle natural movement and a slow camera push-in. Preserve the subject and background, with consistent lighting and smooth cinematic motion.",
"preset": "tuned",
"resolution": "480p",
"duration": 5
}
}'
# 2) Poll until status is "completed", then read the output URLs
curl "https://www.namifusion.com/api/v1/marketplace/run/tasks/TASK_UUID" \
-H "Authorization: Bearer YOUR_API_KEY"Python
import time, requests
API_KEY = "YOUR_API_KEY"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}
# 1) Submit
resp = requests.post(
"https://www.namifusion.com/api/v1/marketplace/run/lightricks/ltx-2.3-spicy/image-to-video-lora",
headers=HEADERS,
json={
"input": {
"image": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
],
"prompt": "Bring the reference image to life with subtle natural movement and a slow camera push-in. Preserve the subject and background, with consistent lighting and smooth cinematic motion.",
"preset": "tuned",
"resolution": "480p",
"duration": 5
}
},
)
resp.raise_for_status() # 401/402/429/5xx stop here instead of polling a bad task
task = resp.json()
# 2) Poll until a terminal state (completed / failed / cancelled).
# This model is allowed up to 300s server-side.
deadline = time.time() + 360
while task.get("status") not in ("completed", "failed", "cancelled"):
if time.time() > deadline:
raise TimeoutError(f"still {task.get('status')} — keep the task_uuid and poll later")
time.sleep(3)
poll = requests.get(f"https://www.namifusion.com/api/v1/marketplace/run/tasks/{task['task_uuid']}", headers=HEADERS)
poll.raise_for_status()
task = poll.json()
print(task["status"], task.get("output"))JavaScript
const API_KEY = "YOUR_API_KEY";
const HEADERS = { Authorization: `Bearer ${API_KEY}` };
// 1) Submit
const resp = await fetch("https://www.namifusion.com/api/v1/marketplace/run/lightricks/ltx-2.3-spicy/image-to-video-lora", {
method: "POST",
headers: { ...HEADERS, "Content-Type": "application/json" },
body: JSON.stringify({
"input": {
"image": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/638e1ee18ac7.jpeg"
],
"prompt": "Bring the reference image to life with subtle natural movement and a slow camera push-in. Preserve the subject and background, with consistent lighting and smooth cinematic motion.",
"preset": "tuned",
"resolution": "480p",
"duration": 5
}
}),
});
if (!resp.ok) throw new Error(`submit failed: ${resp.status} ${await resp.text()}`);
let task = await resp.json();
// 2) Poll until a terminal state (completed / failed / cancelled).
// This model is allowed up to 300s server-side.
const deadline = Date.now() + 360 * 1000;
while (!["completed", "failed", "cancelled"].includes(task.status)) {
if (Date.now() > deadline) throw new Error(`still ${task.status} — keep the task_uuid and poll later`);
await new Promise((r) => setTimeout(r, 3000));
const poll = await fetch(`https://www.namifusion.com/api/v1/marketplace/run/tasks/${task.task_uuid}`, { headers: HEADERS });
if (!poll.ok) throw new Error(`poll failed: ${poll.status}`);
task = await poll.json();
}
console.log(task.status, task.output);Documentation
ai/ltx-2.3-spicy/image-to-video-lora
Fast image-to-video generation with flexible LoRA styling, no cold starts, and cost-efficient output.
ai/ltx-2.3-spicy/image-to-video-lora transforms a reference image and prompt into expressive AI video clips with selectable LoRA presets and layered LoRA control. Built for fast inference on MaaS, it offers no cold starts, practical duration and resolution options, and affordable pricing for production and experimentation alike.
🚀 Key Features
- Image-to-video generation: Uses a reference image as the first-frame foundation and animates it based on your prompt.
- Selectable LoRA presets: Choose between
tunedandoriginalto balance general-purpose quality and lighter styling. - Layered LoRA control: Apply up to 3 LoRAs for more customized visual behavior and style control.
- Multiple resolution tiers: Generate in 480p, 720p, or 1080p depending on quality and budget requirements.
- Flexible duration settings: Supports 3 to 20 seconds of output for short-form content, promos, and creative tests.
- No cold starts: Optimized for responsive inference on MaaS, making it suitable for repeatable production workloads.
🛠️ Technical Specifications
| Item | Details |
|---|---|
| Model architecture | LTX 2.3 Spicy image-to-video LoRA model |
| Task type | Image-to-video |
| Input | Image(JPG/PNG recommended) + Prompt |
| First-frame conditioning | Reference image used as the starting frame |
| LoRA preset options | tuned(recommended general-purpose stack), original(lighter styling) |
| Max LoRAs | Up to 3 |
| Resolution | 480p, 720p, 1080p |
| Duration | 3–20 seconds(default: 5) |
| Seed | Supported; -1 uses a random Seed |
| Output Format | Video |
| Latency | Fast inference, no cold start |
| Typical usage | Stylized clips, product videos, character animation, social content |
Sample Prompts
- A fashion model standing on a city street, hair and coat moving gently in the wind, slow camera push-in, cinematic lighting.
- A robotic cat in a neon-lit room, glowing eyes, subtle head movement, flickering screens in the background.
- A premium skincare bottle in a clean studio setup, smooth orbit camera motion, soft reflections, high-end commercial look.
💰 Pricing
| Item | Price |
|---|---|
| Base Price | $0.15 per run |
💡 Best Use Cases
- E-commerce and advertising: Turn product images into motion-rich promotional clips for storefronts, ads, and landing pages.
- Social media content: Create short, visually engaging videos for TikTok, Instagram Reels, and other vertical content channels.
- Character and IP animation: Animate portraits, concept art, or branded characters into expressive short scenes.
- Creative prototyping: Quickly test motion direction, scene mood, and visual style before full production.
🔗 Related Models
- LTX 2.3 Spicy series: This variant focuses on image-to-video generation with LoRA-based style control for more customizable workflows.
tunedvsoriginalpresets:tunedis the recommended default for most use cases, whileoriginalis better for lighter stylistic intervention.
Related models
xAI Grok Imagine Video v1.5 Image to Video
Animate one input image with a text prompt into a 1-15 second video at 480p or 720p.
vidu/q3/image-to-video-spicy
Vidu Q3 Image-to-Video Spicy generates unlimited high-quality videos from images with smooth animations and diverse motion, optimized for scalable content generation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Vidu Q3 Image To Video
Vidu Q3 Image-to-Video turns text prompts into high-quality videos with exceptional visual fidelity and diverse motion. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

MiniMax H3 Spicy Image to Video
Animate a first-frame image with optional prompt and last-frame guidance. Supports 480p, 540p, 768p and 1080p video, with durations from 3 to 15 seconds.
MiniMax H3 Image to Video
MiniMax H3 Image to Video animates a first-frame image into a coherent 2K video, with natural-language motion instructions and optional last-frame control for consistent motion, scene continuity, and cinematic video generation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
ltx-2.5/image-to-video
LTX 2.5 Image-to-Video animates a first-frame image into high-fidelity synchronized audio-video content, with optional last-frame guidance and 720P / 1080P / 2K / 4K output for cinematic videos, social content, ads, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
ai/ltx-2.3-spicy/image-to-video
LTX 2.3 Spicy Image-to-Video API generates expressive AI videos from a reference image and prompt. Choose style presets, duration, and resolution with fast REST inference, no cold starts, and affordable pricing on MaaS.
Kling Omni Video O3 Image-To-Video
Kling Omni Video O3 Image-to-Video transforms static images into dynamic cinematic videos using MVL (Multi-modal Visual Language) technology. Maintains subject consistency while adding natural motion, physics simulation, and seamless scene dynamics. Supports audio generation. Ready-to-use REST API, best performance, no coldstarts, affordable pricing.