MiniMax H3 Text-to-Video API
Make videos with MiniMax H3 Text-to-Video through MartLLM: $0.084 / sec. No MartLLM fee on requests. Adding credit by card costs 5.5% + $0.50.
Show more
Made by MiniMax. Requests are served by third-party providers. Here, sellers serve it through their own provider accounts.
MiniMax H3 Text-to-Video pricing
per second of video| Resolution | List price | Off | You pay |
|---|---|---|---|
| 768p | $0.10 | 16% | lowest$0.084 |
| 2k | $0.16 | 16% | $0.1366 |
A clip is billed at the price of the resolution you ask for, times its length in seconds. Each request is sent to an available third-party provider for this model and retried if the provider fails before it starts answering. No MartLLM fee on requests. Adding credit by card costs 5.5% + $0.50.
Prices as of 06:00 UTC.
Show more
4 sellers on 1 providers can take a request right now. Each request goes to an available seller and moves to the next one if it fails before answering.
| Provider | Sellers |
|---|---|
| Venice AIVerified | 4 |
Request
curl https://api.martllm.com/v1/video/generations \
-H "Authorization: Bearer $UBERLLM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/venice-minimax-h3-text-to-video",
"prompt": "A traveler in a long coat walks into a neon-lit train station at night, slow tracking shot from behind, rain on the glass roof, reflections on the wet floor, cinematic lighting, the shot ends on the departure board",
"resolution": "768P",
"duration_seconds": 5
}'
# → {"id": "<job id>", "status": "queued", "poll_url": "/v1/video/generations/<job id>"}Set UBERLLM_API_KEY to your key. Ask /v1/video/generations/<job id> every 5 seconds until status is "succeeded", then download the file with your key. A job that fails says why and costs nothing. Full API guide
Options
- Tasks
- Text to video
- Resolutions
- 2K, 768P
- Lengths
- 5 s, 6 s, 7 s, 8 s, 9 s, 10 s, 11 s, 12 s, 13 s, 14 s, 15 s
- Shapes
- 16:9, 21:9, 4:3, 1:1, 3:4, 9:16
- Sound
- Supported
- Billing unit
- per second of video
- Model id
minimax/venice-minimax-h3-text-to-video
Try MiniMax H3 Text-to-Video
Settings
The same request as the settings on the left, ready to copy.
curl https://api.martllm.com/v1/video/generations \
-H "Authorization: Bearer $UBERLLM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/venice-minimax-h3-text-to-video",
"prompt": "",
"resolution": "768P",
"duration_seconds": 5
}'
# → {"id": "<job id>", "status": "queued", "poll_url": "/v1/video/generations/<job id>"}What the MiniMax H3 Text-to-Video API is good at
- Clips up to 15 seconds long in one request.
- Sound made with the video, in the same request.
- Output at 2K, 768P.
- Shapes: 16:9, 21:9, 4:3, 1:1, 3:4, 9:16.
Examples to start from
01
A cinematic shot
A traveler in a long coat walks into a neon-lit train station at night, slow tracking shot from behind, rain on the glass roof, reflections on the wet floor, cinematic lighting, the shot ends on the departure board
Prompt only · 16:9 · 8 s
02
City timelapse
Locked-off timelapse of a city skyline from golden hour to blue hour, clouds drifting, windows lighting up one by one, no camera movement, no warped buildings
Prompt only · 16:9 · 6 s
03
Product shot
A glass perfume bottle turns slowly on a white studio turntable, soft key light from the left, shallow depth of field, clean commercial look, vertical framing for social ads
Prompt only · 9:16 · 5 s
MiniMax H3 Text-to-Video API parameters
| Parameter | Type | What it does |
|---|---|---|
| model | minimax/venice-minimax-h3-text-to-video | Required |
| prompt | string | Required. What to make. |
| resolution | 768P, 2K | Optional |
| duration_seconds | 5 to 15 | Optional. Length in seconds. Left out: the model's shortest. |
| aspect_ratio | 16:9, 21:9, 4:3, 1:1, 3:4, 9:16 | Optional. Shape of the video. |
| audio | boolean | Optional. Add a sound track. |
Read from GET https://api.martllm.com/v1/models/minimax/venice-minimax-h3-text-to-video: the same rules the API checks every request against.
MiniMax models compared
| Model | Model id | From, a second | Up to | Longest |
|---|---|---|---|---|
| MiniMax H3 Text-to-Video | minimax/venice-minimax-h3-text-to-video | $0.084 | 768P | 15 s |
| MiniMax H3 Max Text-to-Video | minimax/venice-minimax-h3-max-text-to-video | $0.0473 | 1080P | 15 s |
| MiniMax H3 Max Image-to-Video | minimax/venice-minimax-h3-max-image-to-video | $0.0473 | 1080P | 15 s |
| MiniMax H3 Enhanced Text-to-Video | minimax/venice-minimax-h3-enhanced-text-to-video | $0.10 | 768P | 15 s |
| MiniMax H3 Image-to-Video | minimax/venice-minimax-h3-image-to-video | $0.10 | 768P | 15 s |
| MiniMax H3 Hyper Text-to-Video | minimax/venice-minimax-h3-hyper-text-to-video | $0.0262 | 768P | 15 s |
| MiniMax H3 Chili Text-to-Video | minimax/venice-minimax-h3-chili-text-to-video | $0.0315 | 768P | 15 s |
| MiniMax H3 Chili Image-to-Video | minimax/venice-minimax-h3-chili-image-to-video | $0.14 | 768P | 15 s |
| MiniMax H3 Hyper Image-to-Video | minimax/venice-minimax-h3-hyper-image-to-video | $0.0276 | 768P | 15 s |
| MiniMax H3 | minimax/minimax-h3 | $0.0543 | 768p | 15 s |
MiniMax H3 Text-to-Video API FAQ
How much does MiniMax H3 Text-to-Video cost per second?
On November 22, 2026, MiniMax H3 Text-to-Video costs from $0.084 per second of video through MartLLM, 48.3% below its list price of $0.16 per second. 4 sellers on the market can serve it. The rate rises with resolution and with sound, and the reply to a request states the most it can cost. A video that fails costs nothing, and MartLLM adds no fee.
How do I make a video with MiniMax H3 Text-to-Video?
Send a POST to https://api.martllm.com/v1/video/generations with your MartLLM key, model "minimax/venice-minimax-h3-text-to-video" and a prompt. You get a job id back; ask https://api.martllm.com/v1/video/generations/<job id> until it says "succeeded", then download it from https://api.martllm.com/v1/media/artifacts/<job id>/0.
Are there other fees?
MartLLM adds no fee to model prices. A card fee of 5.5% + $0.50 is added when you buy credit, and a request that fails is not billed.
Guides
Compare with similar models
Embed this price
Show these prices on your own site. The table reads the current prices from MartLLM, so it stays up to date.
Script: the table sets its own height
<script async src="https://martllm.com/embed.js" data-models="minimax/venice-minimax-h3-text-to-video"></script>Plain iframe, for sites that allow no scripts
<iframe src="https://martllm.com/embed/prices?models=minimax/venice-minimax-h3-text-to-video" title="AI model API prices" width="100%" height="127" loading="lazy" style="border:0;max-width:720px"></iframe>Price badge for a README (Markdown)
[](https://martllm.com/models/minimax/venice-minimax-h3-text-to-video?utm_source=embed&utm_medium=referral&utm_campaign=price-badge)Every option is in the embed docs.