# MiniMax H3 Lite API · KovAPI
> MiniMax H3 Lite is the same MiniMax H3 model run with fewer inference steps: faster and cheaper, with slightly lower quality, made for drafts in bulk and prompt tests. On KovAPI it is `minimax-h3-lite`: 480P or 768P, any length from 4 to 15 seconds, billed per second; for a first frame, a reference video or 1080P, switch to Standard.
Web page: https://kovapi.com/en/model/minimax-h3-lite · Updated: 2026-11-12
Prices, limits and timings are live values from when this file was generated.
## Fact sheet
- Model ID: `minimax-h3` · `minimax-h3-lite`
- Endpoint: `POST https://kovapi.com/v1/videos/generations` → `GET https://kovapi.com/v1/videos/jobs/{id}`
- Length: 4–15 s, whole seconds
- Resolution: 480P / 768P; 768P when omitted; 720P is read as 768P
- Aspect ratio: 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 3:2 / 2:3 / 21:9
- References: up to 5 reference images (references only, never a first frame); no reference video or audio
- Audio: every clip has a native stereo track (dialogue, effects, music); it cannot be turned off
- Prompt: ≤ 5,000 characters
- Per second: 480P $0.015/s · 768P $0.025/s · 1080P $0.036/s
- Lite per second: 480P $0.005/s · 768P $0.0125/s
- Results kept: 7 days
## Picture and sound in one pass, with control over the opening, the ending, the cast and the motion
- **Shots, lines and sound follow the prompt's timeline**: Written shot by shot in the official format, the cut times, the dialogue, the ambience and the music follow the prompt. The bakery clip cuts to the bread knife at 5 seconds, the baker says his line word for word, and the acoustic guitar is the one the prompt asked for.
- **Fix the opening, or both the opening and the ending**: One image as the first frame and the clip starts from it; two images as first and last frame and the model fills in the motion between. The umbrella clip starts on the first picture with the umbrella furled and ends on the second with it open.
- **People and products from your images, combined the way you write**: Reference mode takes up to 9 images. The eyewear ad uses three, the models, their faces and the glasses; written in the reference guide's structure, the models put the glasses on, walk the runway and the clip cuts to a close-up.
- **Borrow the motion and camera of a reference video**: Left, the reference video; right, the result: a cartoon fox dances the same moves as the person. A reference video can also give only the camera move, while the reference image and the prompt decide what is in the shot. Reference seconds are billed too.
- **Draft at 480P, deliver at 1080P**: 480P / 768P / 1080P, eight aspect ratios, any length from 4 to 15 s. Check framing and motion at 480P or on [Lite](https://kovapi.com/en/model/minimax-h3-lite), then render the final clip at 1080P; changing tier renders a new clip, not the same one, and the same-prompt comparison below shows the difference.
## Modes and API fields
Both modes share `model` (`minimax-h3-lite`), `prompt`, `duration`, `resolution` and `size`. Images are always references, so `image_role` is not needed.
| Mode | Send | You get | Limits |
| --- | --- | --- | --- |
| Text to video | Just `prompt` and the generation settings. | Picture and sound from the text; write dialogue, effects and music into the prompt. | Any whole number of seconds from 4 to 15; 768P without `resolution`. |
| Reference images | 1–5 images in `images`. | The people, products or places in them appear in a new shot, never as the first frame. | Up to 5; `image_role: first_frame` is refused. |
## MiniMax H3 pricing
| Resolution | Standard | Lite | MiniMax list price |
| --- | --- | --- | --- |
| 480P | $0.015/s | $0.005/s | - |
| 768P | $0.025/s | $0.0125/s | $0.08 |
| 1080P | $0.036/s | - | - |
- One clip costs the tier's rate × (output seconds + reference-video seconds); reference images and audio are free. Drag "Reference video" in the calculator to price a request with one.
- MiniMax's own price list has H3 at 768P and 2K only, so 768P is the one tier with an official price beside ours.
- The amount is held at submit and settled as is on success; failed jobs are refunded in full. For cheaper drafts in bulk, see [MiniMax H3 Lite](https://kovapi.com/en/model/minimax-h3-lite).
## One prompt, five tiers
Hover over the clips to play them together; on a phone, tap "Play together". Each clip is a separate render, so they are not identical: compare sharpness and detail.
### Latte art: three H3 tiers and two Lite tiers
The same prompt, 5 seconds at 16:9, rendered once each on H3 at 1080P / 768P / 480P and on Lite at 768P / 480P. Check framing and motion at 480P or on Lite, then render the final clip at 1080P.
- 5s · 16:9
- **MiniMax H3 1080P** ($0.18 at current prices): https://img.kovapi.com/i/mp-30a37c4381ec04a2ba543aab.mp4 — Sharpest: the latte pattern, the jug's reflections and the cup rim are all crisp.
- **MiniMax H3 768P** ($0.125 at current prices): https://img.kovapi.com/i/mp-2972abbc7c033fb49b25783d.mp4 — A little less detail than 1080P, still clean.
- **MiniMax H3 480P** ($0.075 at current prices): https://img.kovapi.com/i/mp-faf663f1a053b4a823c2ea31.mp4 — Framing and motion are clear; soft when enlarged. Good for drafts.
- **MiniMax H3 Lite 768P** ($0.0625 at current prices): https://img.kovapi.com/i/mp-4eda7882119d1b49c37bf0e1.mp4 — Fast and cheap; the latte pattern is clearly weaker.
- **MiniMax H3 Lite 480P** ($0.025 at current prices): https://img.kovapi.com/i/mp-10b056d268a7c3beaf4f26e0.mp4 — Cheapest; only for checking framing and motion.
Original prompt:
```text
integrated_multimodal_description: [Shot 1] Live-action, cinematic, a close shot of a barista's hands pouring steamed milk into a cup of espresso on a wooden counter by a sunny window. The milk forms a heart in the crema while steam curls upward, and the camera pushes in with small amplitude at slow speed toward the cup.
overall_soundscape: Milk streams into the cup, a steam wand hisses softly in the background and cafe chatter murmurs.
non_diegetic_music: N/A
```
## Examples: prompts and parameters
### Two shots: the bakery's first batch
- Made with KovAPI · Text to video · 8s · 1080P · 16:9
- $0.288 at current prices
- Output: https://img.kovapi.com/i/mp-c9751f98f2f98ea8114fa1f3.mp4
The text-to-video example in MiniMax's base guide: two shots with a cut to a close-up at 5 seconds. The baker's line, the shutters, the bread knife and the acoustic guitar are all in the prompt, and the clip delivers each. The signs in the background are gibberish; words only come out right when they are quoted in the prompt. Prompt quoted word for word from MiniMax's [Video Prompt Writing Guide](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md); rendered by KovAPI with that prompt.
Original prompt:
```text
integrated_multimodal_description: [Shot 1] Live-action, cinematic, a medium-wide shot frames a baker opening the shutters of a small street bakery before sunrise. The camera pushes in with small amplitude at slow speed as the middle-aged baker with a calm, slightly raspy voice (S1) places a fresh loaf on the wooden counter and says: [English] First batch of the morning. [Shot 2] At 00:05.000, the camera cuts to a close-up of steam rising from the sliced bread while the baker's final words carry over from the previous shot.
overall_soundscape: Wooden shutters scrape open over a quiet street as trays clink softly inside the bakery. The doorbell rings once, followed by light footsteps and the crisp sound of bread being sliced.
non_diegetic_music: A soft acoustic-guitar pattern at a moderate tempo, joined by sparse upright-bass notes and a gentle fade at the end.
```
### Jump to hyperspace on the bridge
- Made with KovAPI · Text to video · 8s · 1080P · 16:9
- $0.288 at current prices
- Output: https://img.kovapi.com/i/mp-66d0f99896d2e402c3af40c3.mp4
The text-to-video example in MiniMax's model repository: a slow push-in on the captain's back, a cut to her face at 4.5 seconds, a blinding flash as the fleet jumps, with the low rumble and score rising together. Prompt quoted word for word from the example in MiniMax's [H3 model repository](https://huggingface.co/MiniMaxAI/MiniMax-H3); rendered by KovAPI.
Original prompt:
```text
integrated_multimodal_description: [Shot 1] Cinematic, medium wide shot, pushing in slowly. In the cavernous, dimly lit bridge of a starship, sleek metallic consoles with glowing amber displays flank a massive, curved observation window. A female captain, in her late 40s with an athletic build and short silver-streaked black hair, stands in the center midground. She wears a structured, high-collared dark navy military tunic with silver chest insignias. Her back is to the camera, silhouetted against the cool, ambient starlight pouring through the thick glass. She stands perfectly still with her hands clasped tightly behind her back. Outside the window, a massive armada of jagged, dark grey dreadnoughts hovers in tight formation against a deep purple space nebula. The fleet's massive rear thrusters begin to glow with an intense, escalating bright blue light. [Shot 2] At 00:04.500, the camera cuts to a close-up of the captain's face and shakes strongly. The brilliant blue-white light from the fleet's gathering energy reflects vividly in her dark eyes. Suddenly, a blinding white flash floods through the window, completely washing out the background as the fleet jumps to hyperspace. The sheer spatial force violently jolts the bridge, causing the captain from Shot 1 to stagger slightly forward, her shoulders tensing as she visibly braces herself against the physical tremors. As the intense white light fades abruptly, leaving only the dim, empty expanse of the purple nebula reflected on her starkly lit skin, her jaw clenches, and she slowly closes her eyes in the newly emptied space.
overall_soundscape: A low, resonant hum of the ship's ambient life support systems serves as the baseline, soon drowned out by an audible, escalating, high-pitched electronic whine as the fleet outside charges its hyperdrives. A massive, deafening, bass-heavy boom and sharp crackle erupts during the blinding flash, accompanied by the loud metallic creaking, rattling, and deep thuds of the bridge's bulkheads vibrating under immense physical stress. The intense roaring impact then cuts abruptly back to a hollow, echoing room tone, leaving only the faint, steady hum of the isolated bridge.
non_diegetic_music: Cinematic space-opera orchestral score, slow tempo, featuring a solitary, mournful French horn melody over deep, sustained string dissonances that build rapidly in volume and intensity, swelling to a massive orchestral peak before snapping immediately into silence right after the jump.
```
### A Chinese line and a neon sign
- Made with KovAPI · Text to video · 6s · 768P · 9:16
- $0.15 at current prices
- Output: https://img.kovapi.com/i/mp-3ea471e559f9efb09edaf873.mp4
Vertical 9:16: the sign text is quoted and the cook's Chinese line is written as `[Chinese] …`. “营业中” and the line are both right; Chinese captions appeared at the bottom of the frame even though the prompt asks for none. Prompt written by KovAPI in the guide's format; rendered by KovAPI.
Original prompt:
```text
integrated_multimodal_description: [Shot 1] Live-action, cinematic, a vertical medium shot of a small street-food stall in a narrow old alley at night. A red neon sign reading "营业中" glows above the stall and steam rises from a wok. The middle-aged cook with a warm, slightly hoarse voice (S1) tosses noodles in the wok as flames flare up, then looks at the camera and says: [Chinese] 刚出锅的，趁热吃！ The camera pushes in with small amplitude at slow speed toward the wok. No subtitles or captions appear on screen.
overall_soundscape: Oil sizzles and the wok clangs against the burner while distant scooters and chatter fill the alley.
non_diegetic_music: N/A
```
### First frame: a letter on a rainy train
- Made with KovAPI · First frame · 6s · 1080P · 16:9 · image_role=first_frame
- $0.216 at current prices
- Output: https://img.kovapi.com/i/mp-94892704d80349abfc4bbf4e.mp4
- Image 1: https://img.kovapi.com/i/mp-71b5ed84bb210183de898f1c.jpg
The first-frame example in MiniMax's base guide: the alignment line comes first, the clip starts on the first frame, she looks up, says her line and folds the letter. Prompt quoted word for word from MiniMax's [Video Prompt Writing Guide](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md); the input images were made by KovAPI with Nano Banana Pro from the prompt's description.
Original prompt:
```text
For the target video, at 0.00 seconds into the target video, (from [Shot 1]) is fully referenced.
integrated_multimodal_description: [Shot 1] Live-action, cinematic, the young woman shown in remains beside the rain-covered train window, preserving her appearance, clothing, seat position, and the carriage layout. The camera trucks right with small amplitude at slow speed as she lifts her gaze from the folded letter toward the passing city lights. Her reflection moves across the glass while the quiet, breathy young woman (S1) says: [English] I get off at the next station. She folds the letter along its existing crease.
overall_soundscape: The train wheels produce a steady metallic rhythm beneath a low ventilation hum. Rain ticks against the window while paper rustles softly in her hands.
non_diegetic_music: Sustained cello notes at a slow tempo with widely spaced piano tones, gradually decreasing in volume.
```
### First + last frame: opening an umbrella in the rain
- Made with KovAPI · First + last frame · 8s · 1080P · 16:9 · image_role=first_frame
- $0.288 at current prices
- Output: https://img.kovapi.com/i/mp-227363d4eedb3bf23a3a199f.mp4
- Image 1: https://img.kovapi.com/i/mp-a560b1808c395d8022f926de.jpg
- Image 2: https://img.kovapi.com/i/mp-c32b3d1c8f0211a33cc0047c.jpg
The first + last frame example in MiniMax's base guide: the two pictures land on second 0 and second 8, and the body only describes the motion between. The clip starts on the furled umbrella and ends on the open one, with the rain and the umbrella's click. Prompt quoted word for word from MiniMax's [Video Prompt Writing Guide](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md); the input images were made by KovAPI with Nano Banana Pro from the prompt's description.
Original prompt:
```text
How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 1) aligns with the 8.00-second mark of the target video.
integrated_multimodal_description: [Shot 1] Live-action, cinematic, a rain-soaked cyclist begins in the position and framing established by Picture 1, holding a closed black umbrella beside a silver bicycle. The camera pulls out with small amplitude at slow speed as she releases the bicycle handle, raises the umbrella above her shoulder, and presses the runner upward until the canopy opens. Water rolls from the expanding fabric while she steps beneath it, rotates the handle into the final angle, and settles into the pose, spacing, and composition established by Picture 2 at the end of the shot.
overall_soundscape: Rain falls steadily on the pavement, followed by the metallic click of the umbrella runner and the soft snap of the canopy opening. Water drips from the bicycle frame as distant traffic passes.
non_diegetic_music: N/A
```
### Reference images: a vertical eyewear ad
- Made with KovAPI · Multi-image reference · 8s · 1080P · 9:16 · image_role=reference
- $0.288 at current prices
- Output: https://img.kovapi.com/i/mp-6a00ea489194c7c6e685fcf7.mp4
- Image 1: https://img.kovapi.com/i/mp-c542946f4a47ea07d3225571.jpg
- Image 2: https://img.kovapi.com/i/mp-dfa785dd4d3f560bb3eb9492.jpg
- Image 3: https://img.kovapi.com/i/mp-802c1b4d4b5ba8f59682ec61.jpg
Three reference images: the two models, their faces and the eyewear design. Written in the reference guide's structure, the models wear the glasses on the runway, the clip cuts to a close-up and adds a product shot at the end. The same idea as a one-liner only showed the three images one after another. The three reference images were made by KovAPI with Nano Banana Pro; the idea is the eyewear ad on MiniMax's [highlights page](https://platform.minimax.io/docs/guides/video-prompt), rewritten by KovAPI in the structure of the [reference guide](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_ref_en.md).
Original prompt:
```text
subject_definitions:
is the two female fashion models in , with slicked-back dark hair, one in a black asymmetric blazer with black-and-silver trousers, the other in a black leather crop top and a long silver satin skirt.
is the facial detail of the same two models in , defining their features and cool expressions.
is the futuristic silver sunglasses in , with wrap-around curves, a sharp cat-eye and goggle hybrid silhouette, mirrored lenses and thin streamlined temples.
summary:
[reference generation] A 9:16 high-fashion eyewear commercial on a minimal white cyc: , with the faces of , wear and walk toward the camera with runway attitude, ending on a close-up of the sunglasses on one model's face.
retention_analysis:
(appears in [Shot 1]): fully_preserved - the two models' outfits and slicked-back hair are retained.
(appears in [Shot 1], [Shot 2]): fully_preserved - their facial features are retained.
(appears in [Shot 1], [Shot 2]): fully_preserved - the wrap-around silver frame and mirrored lenses are retained.
detailed_description:
The target video is a clean, high-fashion commercial with soft, even studio light on a seamless white cyc.
[Shot 1] A vertical full-body shot: both models of , each wearing a pair of , walk side by side toward the camera with a confident runway stride and cool demeanor, their faces matching . The camera pulls out with small amplitude at slow speed.
[Shot 2] At 00:04.000, the camera cuts to a close-up of the left model's face as she turns her head slightly; the mirrored lenses of reflect the white studio, and the camera pushes in with small amplitude at slow speed toward the curved silver frame.
overall_soundscape:
Crisp heel clicks on the studio floor, then a soft whoosh as she turns her head.
non_diegetic_music:
A minimal electronic beat at a moderate tempo with deep synth bass, ending on a single sustained chord.
```
### Reference video: a running shoe on a pedestal
- Made with KovAPI · Video reference · 6s · 768P · 16:9 · image_role=reference
- $0.325 at current prices
- Output: https://img.kovapi.com/i/mp-aa7f83de1a85c8445731122b.mp4
- Image 1: https://img.kovapi.com/i/mp-cc795e889918bc095cfac8c6.jpg
- Reference video (7s): https://img.kovapi.com/i/mp-dd0414d99599d8386d11fbfd.mp4
The reference video circles a statue slowly; the reference image is a running shoe. The clip puts the shoe on the same pedestal in the same square, the camera turns only a little and a cut is added at the end: a camera reference lends the setting and pace, not always the same range of movement. The reference video was also made by KovAPI with H3, the reference image with Nano Banana Pro, and the prompt written by KovAPI.
Original prompt:
```text
The running shoe from image 1 stands on a stone pedestal in the middle of an empty plaza at golden hour. Follow the camera movement and pacing of video 1: a slow arc shot around the subject from left to right, keeping it centered. Do not show the statue from video 1. Sound: soft wind and distant pigeons. No music.
```
### Reference video: a fox dances the same routine
- Made with KovAPI · Video reference · 6s · 768P · 16:9 · image_role=reference
- $0.325 at current prices
- Output: https://img.kovapi.com/i/mp-e2308084597964acbd152bce.mp4
- Image 1: https://img.kovapi.com/i/mp-f75baad27af554d861a38176.jpg
- Reference video (7s): https://img.kovapi.com/i/mp-f34bfc40277ff5ab18021caa.mp4
The reference video is a person dancing; the reference image is a cartoon fox. The fox raises its arms, sways and spins in the same order as the reference. The reference video was also made by KovAPI with H3, the reference image with Nano Banana Pro, and the prompt written by KovAPI.
Original prompt:
```text
The cartoon red fox in the yellow hoodie from image 1 performs the same dance routine as the dancer in video 1, with the same arm raises, hip sways, spin and ending pose, in a colorful cartoon bedroom. 3D animated style, static full-body shot. Sound: upbeat pop drums and bass, the fox's sneakers tapping on the floor.
```
### Lite: a turntable shot of a running shoe
- Made with KovAPI · Multi-image reference · 5s · 768P · 16:9
- $0.0625 at current prices
- Output: https://img.kovapi.com/i/mp-5a3d34dd41ce101dcf3825c9.mp4
- Image 1: https://img.kovapi.com/i/mp-cc795e889918bc095cfac8c6.jpg
Made with [MiniMax H3 Lite](https://kovapi.com/en/model/minimax-h3-lite): one reference image of a running shoe, 768P, 5 seconds. Its cost is at Lite prices, and it fills the playground on the Lite page only. Prompt written by KovAPI in the guide's format; input images made by KovAPI with Nano Banana Pro.
Original prompt:
```text
The running shoe from the reference image rotates slowly on a glossy black turntable while streaks of neon light sweep across it, close studio product shot. Sound: a low electronic hum.
```
## MiniMax H3 prompts: follow the official structure
Based on MiniMax's own Video Prompt Writing Guide ([base](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_base_en.md), [reference mode](https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/docs/VIDEO_PROMPT_WRITING_GUIDE_ref_en.md)). MiniMax says prompt structure matters more on H3 than on most models: write the picture shot by shot, put dialogue in tags, and give ambience and music a section each. The habits below come from that guide and the examples are prompts this page's clips were made with; points marked "tested" are what our own renders showed.
- **Write three sections**: `integrated_multimodal_description` for picture, action, shots and dialogue; `overall_soundscape` for ambience and action sounds in 1–4 sentences; `non_diegetic_music` for music only the audience hears, or `N/A`. Tested: the same bakery scene as a one-liner still got its line and its cut, but no music and the wrong sounds; in three sections the guitar, the doorbell and the bread knife all came out as written. — Bakery, opening words: “integrated_multimodal_description: [Shot 1] Live-action, cinematic, …”
- **One block per shot, with the cut time**: `[Shot 1]` has no time; later shots read `[Shot 2] At 00:05.000, the camera cuts to…`. In the multi-shot clips on this page the cuts landed close to the times written. — Bakery: “[Shot 2] At 00:05.000, the camera cuts to a close-up of steam rising from the sliced bread”
- **Camera = type + amplitude + speed**: Push in, pan, truck, arc shot, tracking shot, plus with small amplitude or at slow speed, written as part of the action rather than tags stacked at the end. — Bakery: “The camera pushes in with small amplitude at slow speed as the middle-aged baker…”
- **Dialogue in its own tags**: Number speakers (S1), (S2), describe age and voice first, then write the line as `[English] …`, word for word and untranslated. Chinese lines use `[Chinese]`. — Train: `the quiet, breathy young woman (S1) says: [English] I get off at the next station.`
- **On-screen words in quotes**: Put the words of a sign or banner in English double quotes exactly as they should appear. Tested: “营业中” came out right, while signs not written into the prompt came out as gibberish. — Noodle stall: “A red neon sign reading "营业中" glows above the stall”
- **Chinese dialogue comes with captions**: Tested: both renders of a Chinese line burned Chinese captions into the bottom of the frame, and asking for no captions did not remove them; the English lines had none. For a clean frame use English dialogue or fix it in post. — Noodle stall, last sentence (it did not work): “No subtitles or captions appear on screen.”
- **Frames: alignment line and path**: A first-frame task opens with the alignment line; first + last says which second the second picture lands on, and the body describes the motion between them instead of describing both pictures again. — Umbrella: “Picture 2 (from Shot 1) aligns with the 8.00-second mark of the target video”
- **Name every reference**: Label each input ``, `` and say what it does in the clip. Tested: the one-line style of MiniMax's highlights page (“Image 1: …, Image 2: …”) turned three reference images into three nearly still shots; rewritten in the reference-guide structure, the models put the glasses on and walked. — Eyewear: ` is the futuristic silver sunglasses in …`
| Part | Purpose | Example |
| --- | --- | --- |
| Picture section | Starts with `integrated_multimodal_description:`, then one block per shot. | Original: “integrated_multimodal_description: [Shot 1] …” |
| Style and framing | The first shot opens with style and shot size. | Original: “Live-action, cinematic, a medium-wide shot frames a baker opening the shutters…” |
| Camera and action | Camera type, amplitude and speed in the same sentence as the action. | Original: “The camera pushes in with small amplitude at slow speed as the middle-aged baker… places a fresh loaf on the wooden counter” |
| Dialogue | Speaker number, voice and the tagged line. | Original: `with a calm, slightly raspy voice (S1) … says: [English] First batch of the morning.` |
| Cut | A new shot gives its cut time and the new picture. | Original: “[Shot 2] At 00:05.000, the camera cuts to a close-up of steam rising from the sliced bread” |
| Two sound sections | One for ambience, one for music. | Original: “overall_soundscape: Wooden shutters scrape open… non_diegetic_music: A soft acoustic-guitar pattern…” |
```text
integrated_multimodal_description: [Shot 1] [style, e.g. Live-action, cinematic], [shot size] frames [subject and setting]. The camera [move] with [small / large] amplitude at [slow / fast] speed as [action]. The [speaker's age and voice] (S1) says: [English] [line] [Shot 2] At 00:0X.000, the camera cuts to [next picture].
overall_soundscape: [ambience and action sounds, 1–4 sentences]
non_diegetic_music: [instruments, tempo and dynamics; N/A for no music]
```
## How to generate video with MiniMax H3 Lite on KovAPI
1. **Sign up and create an API key** Sign in to KovAPI and create a key in the console; one key calls every model on the site.
2. **Try prompts at the top of this page** Pick length, resolution and aspect ratio, add reference images, and see the live price before submitting.
3. **Submit an async job** Send `minimax-h3-lite` to `POST /v1/videos/generations` and keep the job id from the HTTP 202 reply; full request examples are in the docs.
4. **Poll the result or take a webhook** Poll the job id, or set `callback_url` when submitting. On success download the MP4 from `data[0].url`; clips are kept 7 days, so save them in time.
### Minimal request
```bash
curl https://kovapi.com/v1/videos/generations \
-H "Authorization: Bearer $KOVAPI_KEY" -H "Content-Type: application/json" \
-d '{"model": "minimax-h3-lite", "prompt": "...", "duration": 4, "resolution": "1080P", "size": "16:9"}'
# -> HTTP 202 {"id": "..."}; poll:
curl https://kovapi.com/v1/videos/jobs/ -H "Authorization: Bearer $KOVAPI_KEY"
```
## How MiniMax's video models evolved
- 2025-02-11 · T2V-01-Director: **T2V-01-Director · camera control** Camera instructions written in the prompt, and the model moves the camera as told; H3's prompt guide keeps this way of writing camera moves.
- 2025-06-18 · MiniMax Hailuo 02: **Hailuo 02 · 1080P** The second-generation video model, with 1080P and clips up to 10 seconds.
- 2025-10-28 · MiniMax Hailuo 2.3: **Hailuo 2.3 / 2.3 Fast** Better body movement, facial expression, physical realism and prompt adherence; the Fast version is image-to-video only.
- 2026-07-31 · MiniMax H3: **MiniMax H3 · third generation** Text, images, video and audio understood together, picture and stereo sound in one pass; first frame, first + last frame, reference images and reference video in one model. KovAPI offers 480P / 768P / 1080P.
## MiniMax H3 or another video model?
| Model | Length (s) | Resolution | Refs image / video / audio | Price | Good for |
| --- | --- | --- | --- | --- | --- |
| MiniMax H3 Lite (`minimax-h3-lite`) | 4–15 | 480P / 768P | 5/0/0 | $0.005/s | Bulk drafts and prompt tests: the same H3 with fewer inference steps, faster and cheaper with slightly lower quality; text and reference images only. |
| MiniMax H3 (`minimax-h3`) | 4–15 | 480P / 768P / 1080P | 9/3/3 | $0.015/s | 1080P clips with sound, first + last frames, or reference images with a reference video; cheap 480P drafts too. |
| Wan 3.0 (`wan3.0-video`) | 4–30 | 480P / 720P / 1080P | 10/5/5 | $0.04/s | Clips longer than 15 s (up to 30), or video from a document or a web page. |
| Seedance 2.5 (`seedance-2.5`) | 4–30 | 480P / 720P / 1080P | 30/10/10 | $0.10/s | Lots of reference material, or story clips up to 30 s. |
| Seedance 2.0 Fast (`seedance-2.0-fast`) | 5–15 | 480P / 720P | 9/3/3 | $0.046/s | Fast, low-cost Seedance clips. |
| Gemini Omni 1.2 Flash (`gemini-omni-1.2-flash`) | 3–15 | 720P / 1080P / 4K | 3/0/0 | $0.035/s | Editing an existing video with one sentence, 4K, or exact on-screen text. |
| Veo 3.1 (`veo-3.1`) | 8 | 720P / 1080P | 3/0/0 | $0.20/clip | Fixed 8-second clips billed per clip. |
| Grok Imagine Video 1.5 (`grok-imagine-video-1.5`) | 4–15 | 480P / 720P | 7/0/0 | $0.30/clip | 480P / 720P clips billed per clip, with up to 7 reference images. |
If you already have clips from another model, run the same inputs and acceptance checks on both and compare picture, motion, sound, render time and real cost before switching. For H3 versus Lite, see the same-prompt comparison above.
## Combination rules
- Lite has no first frame or first + last frame: images are always references, and `image_role: first_frame` is refused at once and not billed. To open a clip on a given image, use Standard.
- Lite takes no reference video or reference audio; a request with either is refused and not billed. Use Standard for those.
- Only 480P and 768P: `1080P` is refused and `720P` is read as 768P. Every clip has sound; for no music write `non_diegetic_music: N/A` in the prompt.
## Why use MiniMax H3 Lite on KovAPI
- **Cheap drafts**: Lite costs $0.005/s at 480P and $0.0125/s at 768P. Try framing and motion for a prompt on Lite in bulk, then switch the chosen ones to Standard at 1080P. The playground shows the price of a request before you submit.
- **Failures are free and retried**: If a job fails, the amount held is refunded in full; when the upstream is busy or errors out, the system retries for you, so you need no retry logic of your own.
- **No concurrency cap**: No account has a limit on parallel jobs, so submit batches straight away; at peak times jobs queue and the API returns `queue_position` and `eta_seconds`, with a webhook when they finish.
- **One API for every video model**: Submitting, polling and webhooks work the same for every video model on the site; change `model` to send the same prompt to Standard, Wan, Seedance and others.
- **Easy sign-up and payment**: Sign up with email, Google or GitHub; top up by WeChat, card, PayPal or USDT; pay as you go with no monthly fee, and the same key also calls the chat and image models on the site.
## MiniMax H3 API FAQ
### What is MiniMax H3?
A video model MiniMax released in July 2026, which MiniMax calls its third generation after Hailuo 01 and Hailuo 02. It understands text, images, video and audio together, and every clip comes with sound. On KovAPI it is `minimax-h3`, with text to video, first frame, first + last frame, reference images and reference video.
### How long can a clip be?
Any whole number of seconds from 4 to 15; 5 seconds without `duration`. A clip runs a fraction of a second longer than requested (a 5-second job is about 5.2 s) and is billed for the seconds requested.
### Which resolution should I pick, and how big is it?
480P for drafts and bulk, 768P for a balance of detail and price, 1080P for the finest picture; 1080P when `resolution` is omitted. At 16:9, 480P is 864×480, 768P about 1344×768 and 1080P about 1920×1080; a job can differ by a few dozen pixels.
### Can I send 720P?
Yes: it renders and bills as 768P. Other values such as 2K or 4K are refused at submit and not billed.
### How do I switch between frames and references?
Without `image_role`, one image is the first frame and two are the first and last frame. With `image_role: reference` they are reference images, up to 9: their people, products and places appear in the shot instead of being used as frames. Reference video or audio needs reference mode.
### What can a reference video do, and how is it billed?
The clip can follow its motion, camera move or pace: the fox on this page dances the same moves as a person in a reference clip. Reference seconds are billed at the tier's rate together with the output seconds, e.g. 768P with 6 output seconds and a 7-second reference is billed as 13 seconds. Up to 3 clips, each 2–15 s and 15 s in total.
### Does reference audio help, and is it billed?
It is free, up to 3 clips, and must come with reference images or a reference video. MiniMax's guide describes it as a reference for a voice or for music; when we asked a character to speak a new line in the reference voice, the voice did not noticeably change in either test, so do not count on it for a voice.
### What formats and sizes can the references be?
Images: JPG / PNG / WEBP (no GIF), up to 16 MB each, 256–5760 px on each side with an aspect ratio of 0.4–2.5; up to 9 in Standard reference mode, up to 5 on Lite. Reference videos (Standard only): MP4 / MOV with H.264 / H.265 video and AAC / MP3 audio, up to 3, each 2–15 s and 15 s in total, up to 50 MB each. Reference audio (Standard only): WAV / MP3, up to 3, each 2–15 s and 15 s in total, up to 15 MB each. Anything outside these is refused at submit (a wrong codec fails during rendering instead); either way you are not charged.
### Can I turn the sound off?
No, every clip has an audio track and `audio: false` is refused. For no music write `non_diegetic_music: N/A`; for a silent file, remove the track after downloading.
### Can the dialogue be in Chinese or other languages?
Yes. Write the line as `[Chinese] …`; our Chinese line came out word for word, but Chinese captions appeared at the bottom of the frame and asking for none did not remove them. MiniMax says dialogue is stable in 11 languages, including Chinese, English, Japanese, Korean, Spanish and Russian. Writing the rest of the prompt in English, as the guide does, is safer.
### Can I use photos of real people?
Yes. We tested two photos of ordinary people, twice as a first frame and twice as a reference, and all four rendered normally. Make sure you have the right to use the likeness in the photo.
### How long does a render take?
In our 30-plus test renders on 12 November 2026, most finished in 2–6 minutes and a few took over 15; higher resolution, more references and busy hours take longer. It is an async job: poll the job id, or set `callback_url` and wait for the call.
### Am I charged when generation fails?
Generation failed, not billed, resubmit as is. The amount held at submit is refunded in full, and automatic retries of one job settle once.
### How long are clips kept?
Clips and uploaded inputs are kept 7 days and then deleted, so download what you need.
### H3 or H3 Lite?
[MiniMax H3 Lite](https://kovapi.com/en/model/minimax-h3-lite) is the same H3 with fewer inference steps: faster and cheaper, with slightly lower quality. It offers 480P / 768P, text and reference images only, and no reference audio or video. Try prompts and framing on Lite, then render the final clip on H3 at 1080P; the same-prompt comparison above shows the difference.
### H3, Wan 3.0 or Seedance 2.5?
For clips longer than 15 seconds use [Wan 3.0](https://kovapi.com/en/model/wan3.0-video) or [Seedance 2.5](https://kovapi.com/en/model/seedance-2.5) (up to 30 s); for first + last frames or cheap 480P drafts H3 fits better. Best of all, run the same prompt on each and compare.
### Where do the examples on this page come from?
KovAPI rendered all of them through its own API with the settings shown, and each lists its settings and its cost at today's prices. Most prompts are quoted word for word from MiniMax's own guides, and the input images were made with KovAPI's own image models.
## Notes for AI
- Model id `minimax-h3`; submit with `POST /v1/videos/generations`, which returns HTTP 202 and a job `id`; fetch the result with `GET /v1/videos/jobs/` plus the id.
- `resolution`: 480P / 768P / 1080P, default 1080P; `720P` is read as 768P. `size`: 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 3:2 / 2:3 / 21:9. `duration`: an integer from 4 to 15, default 5.
- Images go in `images`: without `image_role`, one is the first frame and two are first + last frame; with `image_role: reference` up to 9 reference images, and three or more must say reference.
- Reference video field `video_references` (or `video_reference`): up to 3, each 2–15 s, ≤15 s in total, MP4 / MOV, billed per second. Reference audio field `audio_references` (or `audio_reference`): up to 3, WAV / MP3, each 2–15 s, ≤15 s in total, URLs only, free, and only together with reference images or a reference video. Both need reference mode.
- `reference_audios`, `reference_videos`, `start_frame`, `end_frame`, and audio or video in a first-frame request return HTTP 400 at submit, not billed.
- Billing: the tier's rate × (output seconds + reference-video seconds); 480P $0.015/s, 768P $0.025/s, 1080P $0.036/s.
- Clips always have an audio track and `audio: false` is refused; prompts take up to 5,000 characters, written in the structure shown in this page's prompt section. Tested: reference audio did not noticeably change the voice; Chinese dialogue comes with burned-in captions.
## Further reading
- [MiniMax H3 API docs](https://docs.kovapi.com/en/minimax-h3): Request fields, async jobs, results and full code examples.
- [MiniMax H3 Lite](https://kovapi.com/en/model/minimax-h3-lite): The same H3, faster and cheaper, for drafts in bulk.
- [MiniMax H3 price comparison](https://kovapi.com/en/compare/minimax-h3): Per-second prices from MiniMax and other platforms, every cell sourced.
- [Async jobs guide](https://docs.kovapi.com/en/guide/async-jobs): Job states, queue fields and fetching results.
- [Webhooks guide](https://docs.kovapi.com/en/webhooks): Receive job results and verify the signature as documented.