MiniMax H3
The H3 look up to 4K, from text, a frame or references. By MiniMax.
minimax/h3/image-to-video
const res = await fetch("https://api.elyum.ai/v1/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.ELYUM_API_KEY}`, "Content-Type": "application/json", Prefer: "wait=55", }, body: JSON.stringify({ "model": "minimax/h3/image-to-video", "prompt": "A slow push-in on a lighthouse at dusk, waves breaking below", "inputs": [ { "role": "first_frame", "url": "https://example.com/reference.jpg" } ] }),});const generation = await res.json();console.log(generation.status, generation.outputs);Result
$0.09 at 480p, $0.13 at 768p, $0.27 at 2K and $0.32 at 4K per second of video. Charged when the file is delivered; a render that fails or is canceled costs nothing.
MiniMax H3 image to video uses the model id minimax/h3/image-to-video. The H3 look up to 4K, from text, a frame or references
1. Calling the API
Endpoint
Send a POST to https://api.elyum.ai/v1/generations with a JSON body matching the parameters below. Every model uses the same endpoint; the model field picks this one.
const res = await fetch("https://api.elyum.ai/v1/generations", { method: "POST", headers: { Authorization: `Bearer ${process.env.ELYUM_API_KEY}`, "Content-Type": "application/json", Prefer: "wait=55", }, body: JSON.stringify({ "model": "minimax/h3/image-to-video", "prompt": "A slow push-in on a lighthouse at dusk, waves breaking below", "inputs": [ { "role": "first_frame", "url": "https://example.com/reference.jpg" } ] }),});const generation = await res.json();console.log(generation.status, generation.outputs);Request parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Yes | — | minimax/h3/image-to-video, or minimax/h3 to let your inputs pick the task. |
prompt | string | Yes | — | Sent to the model as written, never shortened. |
inputs | array | Yes | — | Files the task takes, each with a role and a url or file. See Inputs. |
duration | number | No | 5 | 5 to 15, whole numbers |
resolution | string | No | — | 480P, 768P, 2K, 4K |
aspect_ratio | string | No | — | auto |
max_usd | number | No | — | Refuse the render, with nothing held, if it would cost more than this many dollars. |
metadata | object | No | — | Up to 16 string pairs, echoed back on the generation. |
A value outside these lists is refused with a 400 naming the field, before anything is charged. Nothing is quietly rounded or dropped.
Inputs
| Role | Types | Required | Limits |
|---|---|---|---|
first_frame | image | Yes | — |
last_frame | image | No | — |
Cite an input in the prompt by type and order (@image1, @video1, @audio1), or give it a name and cite @name. An input outside these limits is refused with a 400 naming it, before anything is charged.
2. Authentication
API key
Create a key on Keys with the generate and spend scopes and send it as a bearer token. Keep it on your server; never send it to a browser.
Authorization: Bearer $ELYUM_API_KEY3. Queue
Submit and wait
With Prefer: wait=55 the request holds until the render ends or 55 seconds pass. A finished render answers 200 with its outputs; one still running answers 202 with its id.
Poll and cancel
const headers = { Authorization: `Bearer ${process.env.ELYUM_API_KEY}` }; // Poll until the render ends: succeeded, failed or canceledlet generation = await (await fetch(`https://api.elyum.ai/v1/generations/${id}`, { headers })).json();while (generation.status === "queued" || generation.status === "running") { await new Promise((r) => setTimeout(r, 5000)); generation = await (await fetch(`https://api.elyum.ai/v1/generations/${id}`, { headers })).json();} // Or stop it (a hold is refunded)await fetch(`https://api.elyum.ai/v1/generations/${id}/cancel`, { method: "POST", headers });Output links are signed and last 24 hours: download them to your own storage, or fetch the generation again for fresh ones.
4. Pricing
| Resolution | Price |
|---|---|
| 480p | $0.09 / s |
| 768p | $0.13 / s |
| 2K | $0.27 / s |
| 4K | $0.32 / s |
Send ?dry_run=true for the exact price of a request with nothing held. The API spends your API balance; a render that fails or is canceled costs nothing.
More models by MiniMax
MiniMax H3 task by task, then the lowest price of each other model. What a render costs depends on the resolution and length you choose.
| Model | Price | Resolution | Length |
|---|---|---|---|
| MiniMax H3 Text to video | $0.09 – $0.32 / s | 480P – 4K | 5 – 15 s |
| MiniMax H3 Image to video | $0.09 – $0.32 / s | 480P – 4K | 5 – 15 s |
| MiniMax H3 Reference to video | $0.09 – $0.32 / s | 480P – 4K | 5 – 15 s |
| MiniMax H3 Max TurboText to video · Image to video | from $0.05 / s | 480P – 768P | 5 – 15 s |
| MiniMax H3 MaxText to video · Image to video · Reference to video | from $0.09 / s | 480P – 768P | 5 – 15 s |
| MiniMax ImageText to image · Image edit | $0.02 / image | — | — |