MiniMax H3 Text to Video
minimax/h3/text-to-video
MiniMax H3 Text to Video generates coherent 2K videos from text prompts, with flexible 4-15 second duration and six selectable aspect ratios for cinematic scenes, creative videos, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Examples
Parameters
| Name | Type | Default | Constraints | Description |
|---|---|---|---|---|
| prompt *Prompt | textarea | — | ≤ 7000 chars | Text description of the video scene, action, camera movement, and style. |
| aspect_ratioAspect Ratio | select | 16:9 | 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16 | Output aspect ratio. Adaptive selects a suitable ratio automatically. |
| resolutionResolution | select | 2k | 2k | Output video resolution. |
| durationDuration | slider | 5 | 4 ~ 15 · step 1 | Output video duration in seconds. |
Output fields
| Field | Type | Description |
|---|---|---|
| videos | array<string> | Generated video URLs |
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/minimax/h3/text-to-video" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"aspect_ratio": "16:9",
"resolution": "2k",
"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/minimax/h3/text-to-video",
headers=HEADERS,
json={
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"aspect_ratio": "16:9",
"resolution": "2k",
"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/minimax/h3/text-to-video", {
method: "POST",
headers: { ...HEADERS, "Content-Type": "application/json" },
body: JSON.stringify({
"input": {
"prompt": "A cinematic portrait, dramatic rim light, shallow depth of field",
"aspect_ratio": "16:9",
"resolution": "2k",
"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
MiniMax H3 Text to Video
Affordable 2K text-to-video generation with flexible durations, multiple aspect ratios, and no cold start.
MiniMax H3 Text to Video is a production-ready model that generates coherent 2K videos from natural-language prompts. It is designed for creative and commercial workflows, offering flexible 4–15 second outputs, broad aspect ratio support, and reliable performance for rapid iteration.
🚀 Key Features
- 2K video output: Generate high-resolution videos with a fixed 2K output tier for strong visual clarity.
- Text-to-video generation: Turn natural-language descriptions into videos by specifying scene, action, camera movement, and style.
- Flexible duration control: Choose output lengths from 4 to 15 seconds to match testing, prototyping, or final content needs.
- Multiple aspect ratios: Supports 21:9, 16:9, 4:3, 1:1, 3:4, and 9:16 for cinematic, social, and mobile-first formats.
- No cold start: Well suited for production workflows that require consistent responsiveness.
- Transparent pricing: Per-second pricing makes cost planning straightforward across different video lengths.
🛠️ Technical Specifications
| Item | Details |
|---|---|
| Model Name | MiniMax H3 Text to Video |
| Model ID | minimax/h3/text-to-video |
| Model Architecture | Text-to-video generation model |
| Input Format | Text prompt |
| Output Format | Video |
| Resolution | 2K |
| Duration | 4–15 seconds |
| Aspect Ratio | 21:9, 16:9, 4:3, 1:1, 3:4, 9:16 |
| Default Aspect Ratio | 16:9 |
| Default Resolution | 2K |
| Default Duration | 5 seconds |
| Prompt Length | 1–7000 characters |
| Latency | No cold start; actual inference time varies by prompt complexity and duration |
| Typical Workflows | Cinematic scenes, creative video generation, marketing content, production previsualization |
Sample Prompts
- A cinematic ocean wave at sunrise, highly detailed, golden light breaking through clouds, slow push-in camera movement, epic atmosphere.
- A fashion model walking through a futuristic city street at night, neon reflections on wet pavement, smooth tracking shot, premium commercial style.
- A cozy cabin in a misty forest at dawn, sunlight filtering through trees, slow zoom from wide shot to medium shot, realistic and tranquil mood.
💰 Pricing
| Mode | Price |
|---|---|
| Base Price | $0.13 / second |
| 5-second video | $0.65 |
| 10-second video | $1.30 |
| 15-second video | $1.95 |
💡 Best Use Cases
- Marketing and campaign videos: Create polished concept videos for ads, brand storytelling, and promotional assets.
- Social media content: Produce landscape, square, or vertical videos tailored to different platforms.
- Storyboarding and scene prototyping: Convert written ideas into short motion clips for creative planning and review.
- Visual concept development: Explore camera movement, action, lighting, and mood before full production.
🔗 Related Models
- MiniMax H3 Image-to-Video: Generates high-resolution video from a starting image plus prompt guidance.
- MiniMax H3 Reference-to-Video: Generates high-resolution video using reference media and prompt-based control.
Related models
Pixverse V6 Text to Video
PixVerse V6 delivers a massive leap forward, supporting cinematic ultra-HD quality and deep semantic understanding to precisely transform text prompts into dynamic videos with nuanced lighting and realistic physical interactions. Production-ready REST inference API, zero cold starts, built for professional creation and commercial scaling.
Pixverse C1 Text to Video
PixVerse C1 elevates AI video creation by generating film-grade videos directly from text prompts. It offers flexible durations (1–15s), multiple resolutions up to 1080p, and optional native audio generation. Built for seamless integration, our production-ready REST inference API delivers industry-leading performance, zero cold starts, and highly affordable pricing.
Nami Wan 2.7 T2V Spicy
WAN 2.7 Spicy generates high-quality videos directly from text prompts with smoother motion, stronger visual consistency, and more expressive cinematic results. Supports multi-shot generation and prompt optimization for creative content, advertisements, character animation, and cinematic video production. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.
Kling Omni Video O3 Standard Text-To-Video
Kling Omni Video O3 is Kuaishou's advanced unified multi-modal video model with MVL (Multi-modal Visual Language) technology. Text-to-Video mode generates cinematic videos from text prompts with subject consistency, natural physics simulation, and precise semantic understanding. Supports audio generation. Ready-to-use REST API, best performance, no coldstarts, affordable pricing.
Kling Omni Video O1 Text-To-Video
Kling Omni Video O1 is Kuaishou's first unified multi-modal video model with MVL (Multi-modal Visual Language) technology. Text-to-Video mode generates cinematic videos from text prompts with subject consistency, natural physics simulation, and precise semantic understanding. Ready-to-use REST API, best performance, no coldstarts, affordable pricing.
Kling 3.0
Kling 3.0 delivers high-quality text-to-video generation with smooth motion, cinematic visuals, accurate prompt adherence, and native audio for ready-to-share clips. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.
Kling V2.6 Text to Video API
Kling 2.6 delivers top-tier text-to-video generation with smooth motion, cinematic visuals, strong prompt adherence, and native audio for ready-to-share clips. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.
Veo 3.1 Text to Video
Storytelling Powerhouse. 1080p generation featuring dialogue lip-sync and subject consistency (R2V).