Hailuo 2.3 API

Hailuo 2.3 是聚焦的视频模型，支持文生视频和首帧图生视频。

模型 key

• minimax/hailuo-2.3 — Hailuo 2.3

支持的 mode 与输入

模型	Mode	名称	输入要求
minimax/hailuo-2.3	text_to_video	文生视频	Prompt max 2000 chars
minimax/hailuo-2.3	frames_to_video	首帧生成视频	Prompt max 2000 chars Images start_frame: 1-1 files, 10 MB, formats jpg, png, webp

参数

模型	参数	默认值	可选值	适用 mode
minimax/hailuo-2.3	duration	6s	6s, 10s	全部 mode
minimax/hailuo-2.3	resolution	768p	768p, 1080p	全部 mode

价格

模型	价格
minimax/hailuo-2.3	每秒 2 credits；参数倍率：resolution={"768P":1,"1080p":2}

生产建议

• 显式传入 mode。
• 只使用上表列出的参数值。
• 媒体输入类 mode 请先用 POST /api/v1/files 上传素材。
• 如果页面包含多个变体，请使用你真正需要的模型 key。

示例：文生视频

curl https://buble.ai/api/v1/generations \
  -H "Authorization: Bearer $BUBLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax/hailuo-2.3",
    "mode": "text_to_video",
    "prompt": "A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion",
    "duration": "6s",
    "resolution": "768p"
  }'

示例：素材驱动视频

curl https://buble.ai/api/v1/generations \
  -H "Authorization: Bearer $BUBLE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax/hailuo-2.3",
    "mode": "frames_to_video",
    "prompt": "Animate the source into a polished cinematic motion shot with natural camera movement",
    "start_frame": "https://example.com/start.png",
    "duration": "6s",
    "resolution": "768p"
  }'

SDK 示例

下面的示例使用本页的公开模型 key、mode 和参数值。片段假设你已经按快速开始初始化了 SDK client，并引入了对应语言需要的 SDK 类型。媒体输入类 mode 仍然需要先上传本地文件，再把返回的 URL 放入 image_urls、start_frame、video_urls 等公开字段。

JavaScript / TypeScript

const task = await buble.generations.create({
  model: 'minimax/hailuo-2.3',
  mode: 'text_to_video',
  prompt:
    'A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion',
  duration: '6s',
  resolution: '768p',
});

const result = await buble.generations.wait(task.data.id);
console.log(result.data.result?.videos?.[0]?.url);

Python

task = client.generations.create(
    model="minimax/hailuo-2.3",
    mode="text_to_video",
    prompt="A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion",
    duration="6s",
    resolution="768p",
)

result = client.generations.wait(task["data"]["id"])
print(result["data"]["result"]["videos"][0]["url"])

Go

task, err := client.Generations.Create(ctx, &buble.CreateGenerationRequest{
	Model:  "minimax/hailuo-2.3",
	Mode:   "text_to_video",
	Prompt: "A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion",
	Params: map[string]any{
		"duration": "6s",
		"resolution": "768p",
	},
})
if err != nil {
	return err
}

result, err := client.Generations.Wait(ctx, task.Data.ID)
if err != nil {
	return err
}
fmt.Println(result.Data.Result.Videos[0].URL)

Rust

let task = client.generations().create(
    CreateGenerationRequest::new("minimax/hailuo-2.3")
        .mode("text_to_video")
        .prompt("A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion")
        .param("duration", "6s")?
        .param("resolution", "768p")?,
).await?;

let result = client
    .generations()
    .wait(&task.data.id, WaitOptions::default())
    .await?;

Swift

let task = try await client.generations.create(
    try CreateGenerationRequest(model: "minimax/hailuo-2.3")
        .mode("text_to_video")
        .prompt("A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion")
        .param("duration", "6s")
        .param("resolution", "768p")
)

let result = try await client.generations.wait(task.data.id)
print(result.data.result?.videos?.first?.url.absoluteString ?? "")

Dart / Flutter

final task = await client.generations.create(
  CreateGenerationRequest(
    model: 'minimax/hailuo-2.3',
    mode: 'text_to_video',
    prompt: 'A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion',
  ).withParam('duration', '6s').withParam('resolution', '768p'),
);

final result = await client.generations.wait(task.data.id);
print(result.data.result?.videos.firstOrNull?.url);

Elixir

{:ok, task} =
  Buble.Generations.create(client, %{
    model: "minimax/hailuo-2.3",
    mode: "text_to_video",
    prompt: "A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion",
    duration: "6s",
    resolution: "768p"
  })

{:ok, result} = Buble.Generations.wait(client, task["data"]["id"])
IO.puts(result["data"]["result"]["videos"] |> List.first() |> Map.fetch!("url"))

Java

Envelope<GenerationTask> task = client.generations().create(
        CreateGenerationRequest.builder()
                .model("minimax/hailuo-2.3")
                .mode("text_to_video")
                .prompt("A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion")
                .param("duration", "6s")
                .param("resolution", "768p")
                .build());

Envelope<GenerationTask> result = client.generations().wait(task.getData().getId());

.NET

var task = await client.Generations.CreateAsync(new CreateGenerationRequest
{
    Model = "minimax/hailuo-2.3",
    Mode = "text_to_video",
    Prompt = "A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion"
}.WithParam("duration", "6s").WithParam("resolution", "768p"));

var result = await client.Generations.WaitAsync(task!.Data!.Id!);

PHP

$task = $client->generations()->create(
    CreateGenerationRequest::make(
        model: 'minimax/hailuo-2.3',
        mode: 'text_to_video',
        prompt: 'A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion',
    )
    ->withParam('duration', '6s')
    ->withParam('resolution', '768p')
);

$result = $client->generations()->wait($task['data']['id']);

Ruby

task = client.generations.create(
  model: "minimax/hailuo-2.3",
  mode: "text_to_video",
  prompt: "A cinematic product launch shot with slow camera movement, premium lighting, and realistic motion",
  duration: "6s",
  resolution: "768p"
)

result = client.generations.wait(task.dig("data", "id"))
puts result.dig("data", "result", "videos", 0, "url")