MiniMax H3

A scene in. Video with its own audio out, 768p at 24 fps.

POSTauray-ai/minimax-h3/text-to-video

One of the models behind Auray's apps, under the auray-ai/ namespace.

Its request shape is published below. Whether it runs right now is GET /v1/catalogue's word, with no key needed: its addresses say connected for an address that runs, and a submit to any other is refused before anything is charged — described answers 501 model_not_connected, paused 503 model_disabled.

Category text-to-video · Scope video:write (or auray-ai-video:write on an older key) · Price A flat base per job plus one price per clip for each length rung — up to 5, 10 and 15 seconds — the credits of each in force served by GET /v1/plans as price_keys (video.base, video.fast.<rung>) and quoted when you submit. A clip held by reference pictures is charged the per cent in video.reference_multiplier_pct of its rung's price, rounded up once, and the base as it is; a first frame, and shots cut inside one clip, cost nothing extra. Refunded if the render fails.

#What it can do

t2v — Text → Video
story — Story: up to 9 pictures (optional); 1 track (optional), not taken through the API yet
first-frame — First frame: 1 first frame
video-refs — References: up to 9 pictures
Length: 5 to 15 seconds
Aspect ratios: 16:9, 9:16
Prompt: up to 7,000 characters
Sound: makes its own sound
Priced by: rung (the whole rule is under Price)

The same model also answers at auray-serverless/minimax/h3-0, with its own capabilities and its own price — its page says them.

Find the addresses that do any of these with GET /v1/models?operation=<id>; the openapi.json carries them as data under x-auray-capability.

#Input

Field	Type	Required	Default	Description
`prompt`	string	yes	—	What to film. Documented range: up to 7,000 characters.
`duration_seconds`	integer	no	`5`	Whole seconds; outside 5–15 it is refused with 422 duration_out_of_range, before any charge. Documented range: 5 to 15.
`tier`	string	no	`"fast"`	One configuration. Two earlier tiers, `standard` and `pro`, were retired on 2026-08-27 and 2026-09-10, and both are refused now. One of `"fast"`.
`aspect_ratio`	string	no	`"16:9"`	16:9 or 9:16. The first clip after a few quiet minutes waits about 12 minutes longer to start, and after such a wait the first clip longer than 5 seconds on each canvas may take longer to render than the ones after it; the quoted expected_seconds includes both. One of `"16:9"`, `"9:16"`.
`first_frame_path`	string	no	—	The picture the clip opens on: the `path` that `POST /v1/uploads` with this address answered, sent once its bytes have been PUT. Costs nothing extra. Not with `reference_image_paths`: the model's reference mode takes no first frame, and the pair is refused with 422 start_frame_with_references before any charge. A path from POST /v1/uploads; this platform never fetches a URL you give it.
`reference_image_paths`	string[]	no	—	Up to 9 pictures that hold a subject — a face, a costume, a place — each the `path` an upload answered. The prompt names them in this order as `<Picture 1>`, `<Picture 2>` and so on. A clip with references is charged the reference multiplier, `video.reference_multiplier_pct` in GET /v1/plans' `price_keys`, once over the clip's rung price. A path from POST /v1/uploads; this platform never fetches a URL you give it. Documented range: up to 9.
`seed`	integer	no	—	Leave it out and one is chosen; a finished clip reports the seed it rendered with (`seed` on each video), and sending that number back with the same prompt and settings usually renders the same clip — one made from reference pictures or a first frame may differ in fine detail. Outside the range it is refused with 422 seed_out_of_range, before any charge. Documented range: 0 to 4,294,967,295.
`idempotency_key`	string	no	—	Your own name for this request. Send the same one twice and the second call is answered with the first request rather than charged again. `Idempotency-Key:` as a header is accepted as an alias — that is the spelling a retrying proxy replays, and the one to use when running a model on `auray.run`, where a request that outlives its connection is answered 504 and recovered by replay. Omit it and the platform mints one. 1 to 64 characters.

#Example

curl -X POST https://queue.auray.run/auray-ai/minimax-h3/text-to-video \
  -H "Authorization: Bearer $AURAY_API_KEY" \
  -H "content-type: application/json" \
  -d '{
  "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
}'

import { createClient } from "@auray/client";

const auray = createClient(); // reads AURAY_API_KEY

const result = await auray.subscribe("auray-ai/minimax-h3/text-to-video", {
  input: {
    "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
  },
  onQueueUpdate: (update) => console.log(update.status),
});
console.dir(result, { depth: null });

from auray_client import Client

auray = Client()  # reads AURAY_API_KEY

result = auray.subscribe(
    "auray-ai/minimax-h3/text-to-video",
    input={
        "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
    },
    on_queue_update=lambda update: print(update["status"]),
)
print(result)

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run_model",
    "arguments": {
      "address": "auray-ai/minimax-h3/text-to-video",
      "input": {
        "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
      }
    }
  }
}

const res = await fetch("https://queue.auray.run/auray-ai/minimax-h3/text-to-video", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AURAY_API_KEY}`,
    "content-type": "application/json",
  },
  body: JSON.stringify({
    "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
  }),
});
console.log(await res.json());

import os, requests

r = requests.post(
    "https://queue.auray.run/auray-ai/minimax-h3/text-to-video",
    headers={"Authorization": f"Bearer {os.environ['AURAY_API_KEY']}"},
    json={
        "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
    },
)
r.raise_for_status()
print(r.json())

package main

import (
	"bytes"
	"fmt"
	"io"
	"net/http"
	"os"
)

func main() {
	body := []byte(`{
  "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
}`)
	req, err := http.NewRequest("POST", "https://queue.auray.run/auray-ai/minimax-h3/text-to-video", bytes.NewReader(body))
	if err != nil {
		panic(err)
	}
	req.Header.Set("Authorization", "Bearer "+os.Getenv("AURAY_API_KEY"))
	req.Header.Set("Content-Type", "application/json")

	res, err := http.DefaultClient.Do(req)
	if err != nil {
		panic(err)
	}
	defer res.Body.Close()

	out, _ := io.ReadAll(res.Body)
	fmt.Println(res.Status, string(out))
}

<?php

$ch = curl_init("https://queue.auray.run/auray-ai/minimax-h3/text-to-video");
curl_setopt_array($ch, [
    CURLOPT_CUSTOMREQUEST => "POST",
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer " . getenv("AURAY_API_KEY"),
        "Content-Type: application/json",
    ],
    CURLOPT_POSTFIELDS => json_encode([
        "prompt" => "a brass orrery on a walnut desk, morning light from a window to the left",
    ], JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE),
]);

$response = curl_exec($ch);
echo curl_getinfo($ch, CURLINFO_HTTP_CODE), " ", $response, PHP_EOL;

require "net/http"
require "json"

uri = URI("https://queue.auray.run/auray-ai/minimax-h3/text-to-video")
req = Net::HTTP::Post.new(uri)
req["Authorization"] = "Bearer #{ENV.fetch("AURAY_API_KEY")}"
req["Content-Type"] = "application/json"
req.body = {
  "prompt" => "a brass orrery on a walnut desk, morning light from a window to the left"
}.to_json

res = Net::HTTP.start(uri.hostname, uri.port, use_ssl: true) { |http| http.request(req) }
puts res.code, res.body

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;

public class Main {
    public static void main(String[] args) throws Exception {
        String body = """
            {
              "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
            }""";

        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create("https://queue.auray.run/auray-ai/minimax-h3/text-to-video"))
            .header("Authorization", "Bearer " + System.getenv("AURAY_API_KEY"))
            .header("Content-Type", "application/json")
            .method("POST", HttpRequest.BodyPublishers.ofString(body))
            .build();

        HttpResponse<String> response = HttpClient.newHttpClient()
            .send(request, HttpResponse.BodyHandlers.ofString());
        System.out.println(response.statusCode() + " " + response.body());
    }
}

using System.Net.Http.Headers;
using System.Text;

using var client = new HttpClient();
client.DefaultRequestHeaders.Authorization =
    new AuthenticationHeaderValue("Bearer", Environment.GetEnvironmentVariable("AURAY_API_KEY"));

using var request = new HttpRequestMessage(HttpMethod.Post, "https://queue.auray.run/auray-ai/minimax-h3/text-to-video");
var body = """
    {
      "prompt": "a brass orrery on a walnut desk, morning light from a window to the left"
    }
    """;
request.Content = new StringContent(body, Encoding.UTF8, "application/json");

using var response = await client.SendAsync(request);
Console.WriteLine($"{(int)response.StatusCode} {await response.Content.ReadAsStringAsync()}");

This model cannot be waited on: https://auray.run answers 400 model_not_synchronous and ?wait= is refused wait_unsupported, both before anything is charged. Submit it here and poll, stream its status, or pass ?webhook= — the options are on Submit a request.

#Storyboard

A storyboard: 2 to 16 scenes in segments, 16 to 120 seconds in all, filmed one after the other into one film and held by up to 9 pictures across every scene. A body naming segments is a storyboard and takes only these fields — a clip's field beside them, a scene's extra, a track or a music bed is refused storyboard_field_not_taken before anything is charged. Priced per scene at the clip's rungs, the reference multiplier on every scene when pictures hold the cast; the submit's answer carries the whole as credits_quoted, and a read of the job what was taken as credits_charged. It does not wait in a queue: while the platform is full it is refused 503 service_busy with nothing charged — submit again later.

Field	Type	Required	Default	Description
`segments`	object[]	yes	—
`segments[].prompt`	string	yes	—	What happens in this scene. A storyboard is 2 to 16 scenes, filmed one after the other into one film; a scene's prompt names the pictures it shows as `<Picture 1>`, `<Picture 2>` and so on, in the order `reference_image_paths` sends them. Documented range: up to 7,000 characters.
`segments[].duration_seconds`	integer	no	`8`	Whole seconds of this scene, 5 to 15; the scenes together run 16 to 120 seconds. Outside either it is refused before any charge. Documented range: 5 to 15.
`tier`	string	no	`"fast"`	One configuration, as a clip's. One of `"fast"`.
`aspect_ratio`	string	no	`"16:9"`	The film's shape, every scene alike. One of `"16:9"`, `"9:16"`.
`reference_image_paths`	string[]	no	—	Up to 9 pictures that hold the cast across every scene, each the `path` `POST /v1/uploads` with this address answered. A storyboard held by pictures is charged the reference multiplier on every scene, `video.reference_multiplier_pct` in GET /v1/plans' `price_keys`. A path from POST /v1/uploads; this platform never fetches a URL you give it. Documented range: up to 9.
`idempotency_key`	string	no	—	Your own name for this request. Send the same one twice and the second call is answered with the first request rather than charged again. `Idempotency-Key:` as a header is accepted as an alias — that is the spelling a retrying proxy replays, and the one to use when running a model on `auray.run`, where a request that outlives its connection is answered 504 and recovered by replay. Omit it and the platform mints one. 1 to 64 characters.

curl -X POST https://queue.auray.run/auray-ai/minimax-h3/text-to-video \
  -H "Authorization: Bearer $AURAY_API_KEY" \
  -H "content-type: application/json" \
  -d '{
  "segments": [
    {
      "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
      "duration_seconds": 8
    },
    {
      "prompt": "The lamp turns and its beam sweeps across the water",
      "duration_seconds": 8
    }
  ],
  "aspect_ratio": "16:9"
}'

import { createClient } from "@auray/client";

const auray = createClient(); // reads AURAY_API_KEY

const result = await auray.subscribe("auray-ai/minimax-h3/text-to-video", {
  input: {
    "segments": [
      {
        "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
        "duration_seconds": 8
      },
      {
        "prompt": "The lamp turns and its beam sweeps across the water",
        "duration_seconds": 8
      }
    ],
    "aspect_ratio": "16:9"
  },
  onQueueUpdate: (update) => console.log(update.status),
});
console.dir(result, { depth: null });

from auray_client import Client

auray = Client()  # reads AURAY_API_KEY

result = auray.subscribe(
    "auray-ai/minimax-h3/text-to-video",
    input={
        "segments": [
            {
                "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
                "duration_seconds": 8
            },
            {
                "prompt": "The lamp turns and its beam sweeps across the water",
                "duration_seconds": 8
            }
        ],
        "aspect_ratio": "16:9"
    },
    on_queue_update=lambda update: print(update["status"]),
)
print(result)

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run_model",
    "arguments": {
      "address": "auray-ai/minimax-h3/text-to-video",
      "input": {
        "segments": [
          {
            "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
            "duration_seconds": 8
          },
          {
            "prompt": "The lamp turns and its beam sweeps across the water",
            "duration_seconds": 8
          }
        ],
        "aspect_ratio": "16:9"
      }
    }
  }
}

const res = await fetch("https://queue.auray.run/auray-ai/minimax-h3/text-to-video", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AURAY_API_KEY}`,
    "content-type": "application/json",
  },
  body: JSON.stringify({
    "segments": [
      {
        "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
        "duration_seconds": 8
      },
      {
        "prompt": "The lamp turns and its beam sweeps across the water",
        "duration_seconds": 8
      }
    ],
    "aspect_ratio": "16:9"
  }),
});
console.log(await res.json());

import os, requests

r = requests.post(
    "https://queue.auray.run/auray-ai/minimax-h3/text-to-video",
    headers={"Authorization": f"Bearer {os.environ['AURAY_API_KEY']}"},
    json={
        "segments": [
            {
                "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
                "duration_seconds": 8
            },
            {
                "prompt": "The lamp turns and its beam sweeps across the water",
                "duration_seconds": 8
            }
        ],
        "aspect_ratio": "16:9"
    },
)
r.raise_for_status()
print(r.json())

package main

import (
	"bytes"
	"fmt"
	"io"
	"net/http"
	"os"
)

func main() {
	body := []byte(`{
  "segments": [
    {
      "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
      "duration_seconds": 8
    },
    {
      "prompt": "The lamp turns and its beam sweeps across the water",
      "duration_seconds": 8
    }
  ],
  "aspect_ratio": "16:9"
}`)
	req, err := http.NewRequest("POST", "https://queue.auray.run/auray-ai/minimax-h3/text-to-video", bytes.NewReader(body))
	if err != nil {
		panic(err)
	}
	req.Header.Set("Authorization", "Bearer "+os.Getenv("AURAY_API_KEY"))
	req.Header.Set("Content-Type", "application/json")

	res, err := http.DefaultClient.Do(req)
	if err != nil {
		panic(err)
	}
	defer res.Body.Close()

	out, _ := io.ReadAll(res.Body)
	fmt.Println(res.Status, string(out))
}

<?php

$ch = curl_init("https://queue.auray.run/auray-ai/minimax-h3/text-to-video");
curl_setopt_array($ch, [
    CURLOPT_CUSTOMREQUEST => "POST",
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer " . getenv("AURAY_API_KEY"),
        "Content-Type: application/json",
    ],
    CURLOPT_POSTFIELDS => json_encode([
        "segments" => [
            [
                "prompt" => "A lighthouse on a cliff at dusk, the sea calm below",
                "duration_seconds" => 8,
            ],
            [
                "prompt" => "The lamp turns and its beam sweeps across the water",
                "duration_seconds" => 8,
            ],
        ],
        "aspect_ratio" => "16:9",
    ], JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE),
]);

$response = curl_exec($ch);
echo curl_getinfo($ch, CURLINFO_HTTP_CODE), " ", $response, PHP_EOL;

require "net/http"
require "json"

uri = URI("https://queue.auray.run/auray-ai/minimax-h3/text-to-video")
req = Net::HTTP::Post.new(uri)
req["Authorization"] = "Bearer #{ENV.fetch("AURAY_API_KEY")}"
req["Content-Type"] = "application/json"
req.body = {
  "segments" => [
    {
      "prompt" => "A lighthouse on a cliff at dusk, the sea calm below",
      "duration_seconds" => 8
    },
    {
      "prompt" => "The lamp turns and its beam sweeps across the water",
      "duration_seconds" => 8
    }
  ],
  "aspect_ratio" => "16:9"
}.to_json

res = Net::HTTP.start(uri.hostname, uri.port, use_ssl: true) { |http| http.request(req) }
puts res.code, res.body

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;

public class Main {
    public static void main(String[] args) throws Exception {
        String body = """
            {
              "segments": [
                {
                  "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
                  "duration_seconds": 8
                },
                {
                  "prompt": "The lamp turns and its beam sweeps across the water",
                  "duration_seconds": 8
                }
              ],
              "aspect_ratio": "16:9"
            }""";

        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create("https://queue.auray.run/auray-ai/minimax-h3/text-to-video"))
            .header("Authorization", "Bearer " + System.getenv("AURAY_API_KEY"))
            .header("Content-Type", "application/json")
            .method("POST", HttpRequest.BodyPublishers.ofString(body))
            .build();

        HttpResponse<String> response = HttpClient.newHttpClient()
            .send(request, HttpResponse.BodyHandlers.ofString());
        System.out.println(response.statusCode() + " " + response.body());
    }
}

using System.Net.Http.Headers;
using System.Text;

using var client = new HttpClient();
client.DefaultRequestHeaders.Authorization =
    new AuthenticationHeaderValue("Bearer", Environment.GetEnvironmentVariable("AURAY_API_KEY"));

using var request = new HttpRequestMessage(HttpMethod.Post, "https://queue.auray.run/auray-ai/minimax-h3/text-to-video");
var body = """
    {
      "segments": [
        {
          "prompt": "A lighthouse on a cliff at dusk, the sea calm below",
          "duration_seconds": 8
        },
        {
          "prompt": "The lamp turns and its beam sweeps across the water",
          "duration_seconds": 8
        }
      ],
      "aspect_ratio": "16:9"
    }
    """;
request.Content = new StringContent(body, Encoding.UTF8, "application/json");

using var response = await client.SendAsync(request);
Console.WriteLine($"{(int)response.StatusCode} {await response.Content.ReadAsStringAsync()}");

#Machine contracts

openapi.json — this page as an OpenAPI 3.1 document, public.
llms.txt — the same, written for an agent.