跳到主要内容

KlingVideo model检查中

Kling O3 API

模型 IDkling-o3-4k-text-to-video

Kling 3.0 Omni video generation and editing in Standard, Pro and 4K variants.

  • 文生视频
  • 图生视频
  • Reference to video
  • Video edit
  • Up to 7 input images
  • Up to 15 seconds
  • 声音

在 Playground 中运行 Kling O3

The form is the request body: every field shows its key. Sign in to run it; a run is billed like an API task.

参数

模型
model

Kling O34K text to videokling-o3-4k-text-to-video

必填
input.prompt

Text description of the scene, action, camera style, lighting, and mood. ≤ 2500 chars

input.duration

Clip length in seconds (3-15). 3–15

常用参数
input.aspectRatio

Output aspect ratio.

input.shotType

Shot type.When the multiShot parameter is true, the current parameter is required.

高级参数 (4)
input.sound

Whether to generate synchronized audio.

input.multiPrompt

List of multi-prompt segments (optional). ≤ 6 items

input.elementList

Element reference list (optional). ≤ 3 items

input.multiShot

Whether to generate multi-shot video (when true, the prompt parameter is invalid; when false, the shotType parameter and the multiPrompt parameter are invalid)

预估费用
正在加载价格…

任务接收时预留额度,仅成功后扣费。失败任务释放预留额度;结果未确认时会保留额度,直到核实。

请求预览

{"model": "kling-o3-4k-text-to-video","input": {"aspectRatio": "16:9","duration": 5,"sound": false,"shotType": "customize","multiShot": false}}

网页 Playground 使用你的登录会话发送此请求体。在代码中请使用 API Key 发送相同内容。

结果

运行请求后,结果会显示在这里,并附上状态和费用。

Kling O3 价格

各可用版本和规格的价格以美元显示,计费单位随费率注明;只有找到可匹配的官方标价时才显示参考价格。

正在加载价格…
  • Charged only on success仅对成功任务扣费。失败任务释放预留额度;结果尚未确认的任务会保留额度,直到结果核实。
  • Safe retries使用相同幂等密钥重试不会重复扣费。
  • Price changes发送 expectedPriceMicros:若展示后价格发生变化,任务会以 PRICE_CHANGED 被拒绝。

对比 Kling O3 各版本

每个变体都有独立的模型 ID 和输入参数。请先选择任务和版本;变体存在差异时会显示额外字段。

  • Standard text to video默认值

    kling-o3-std-text-to-video

    文生视频; 文本; 3–15 s; 1:1, 16:9, 9:16

  • Pro text to video

    kling-o3-pro-text-to-video

    文生视频; 文本; 3–15 s; 1:1, 16:9, 9:16

    elementList
  • Standard image to video

    kling-o3-std-image-to-video

    图生视频; 文本, 图像; 3–15 s

  • Pro image to video

    kling-o3-pro-image-to-video

    图生视频; 文本, 图像; 3–15 s

    elementList
  • 4K text to video

    kling-o3-4k-text-to-video

    文生视频; 文本; 3–15 s; 16:9, 9:16, 1:1

    elementList
  • 4K image to video

    kling-o3-4k-image-to-video

    图生视频; 文本, 图像; 3–15 s

    elementList
  • 4K reference to video

    kling-o3-4k-reference-to-video

    Reference to video; 文本, 图像; 3–15 s; 16:9, 9:16, 1:1

    elementList
  • Pro video edit

    kling-o3-pro-video-edit

    Video edit; 文本, 图像, 视频

    elementList
  • Standard video edit

    kling-o3-std-video-edit

    Video edit; 文本, 图像, 视频

  • Pro reference to video

    kling-o3-pro-reference-to-video

    Reference to video; 文本, 图像, 视频; 3–15 s; 16:9, 9:16, 1:1

    videoUrlelementList
  • Standard reference to video

    kling-o3-std-reference-to-video

    Reference to video; 文本, 图像, 视频; 3–15 s; 16:9, 9:16, 1:1

    videoUrl

Kling O3 能做什么

Kling describes VIDEO 3.0 Omni as a multimodal video model with native audio and storyboard controls. The manufacturer distinguishes it from the earlier O1 model. Hermes groups O3 task IDs on this page; exact fields vary by tier and task.

The Hermes catalog lists text-to-video, image-to-video, reference-to-video and video-edit variants. Generation variants list a sound control and support structured multi-shot fields on compatible schemas; editing uses a source video and separate sound-preservation options. Some Pro and 4K variants list element references. Check the individual task's limits, status and price.

能力

  • Up to 15 seconds

    Choose a length from 3 to 15 seconds.

    duration

    Every variant except Pro video edit, Standard video edit

  • Multi-shot scenes

    Describe up to 6 shots in one request, each with its own prompt and duration.

    multiShotmultiPrompt

    Every variant except Pro video edit, Standard video edit

  • Start and end frames

    Set the first frame and, optionally, the last frame the clip moves to.

    firstImageUrllastImageUrl

    版本: Standard image to video, Pro image to video, 4K image to video

  • 声音

    Generate audio together with the video.

    sound

    Every variant except Pro video edit, Standard video edit

  • 3 aspect ratios

    1:1, 16:9 or 9:16.

    aspectRatio

    Every variant except Standard image to video, Pro image to video, 4K image to video, Pro video edit, Standard video edit

  • Up to 7 input images

    Send up to 7 images in one request as HTTPS URLs or uploaded media references.

    imageUrls

    版本: 4K reference to video, Pro video edit, Standard video edit, Pro reference to video, Standard reference to video

  • Video input

    Accepts a source video as input.

    videoUrl

    版本: Pro video edit, Standard video edit, Pro reference to video, Standard reference to video

  • Element references

    Add up to 4 element references to the request.

    elementList

    Every variant except Standard text to video, Standard image to video, Standard video edit, Standard reference to video

使用场景

  1. Develop a storyboard

    Describe a sequence with the multi-shot fields supported by the chosen generation variant.

  2. Animate an image

    Supply a first frame and, where supported, an end frame.

  3. Revise an existing clip

    Select video editing and provide its source media.

用代码调用 Kling O3

Every image, video and music model uses the same asynchronous job API: submit a task, then poll for the result.

  1. 1

    创建 API Key

    In the dashboard, with the jobs:create and jobs:read scopes. API 密钥
  2. 2

    提交任务

    POST /api/v1/jobs with the model ID and an input object. The response is 202 with a task ID.
  3. 3

    Poll for the result

    GET /api/v1/jobs/{id} until the status is succeeded, then download the output URLs.

最小请求

Required fields only, generated from the schema of 4K text to video. Set HERMES_API_KEY in your environment.

模型 ID
kling-o3-4k-text-to-video
端点
https://api.hermes-ai.net/api/v1/jobs
Full task API reference
# 1. Submit the task (asynchronous; returns 202 with a task id)
curl -X POST https://api.hermes-ai.net/api/v1/jobs \
  -H "Authorization: Bearer $HERMES_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
  "model": "kling-o3-4k-text-to-video",
  "input": {
    "prompt": "A lighthouse on a rocky coast at sunrise, soft light",
    "duration": 5
  }
}'

# 2. Poll until status is succeeded, failed or reconciliation_required
curl https://api.hermes-ai.net/api/v1/jobs/TASK_ID \
  -H "Authorization: Bearer $HERMES_API_KEY"

输入参数

Fields of 4K text to video (kling-o3-4k-text-to-video). 未知字段会被拒绝并返回 INVALID_MODEL_INPUT。

字段类型必填默认值可选值与约束
prompt
string必填—

Text description of the scene, action, camera style, lighting, and mood.

≤ 2500 chars

aspectRatio
string可选"16:9"

Output aspect ratio.

可选值: "16:9", "9:16", "1:1"

duration
integer必填5

Clip length in seconds (3-15).

3–15

sound
boolean可选false

Whether to generate synchronized audio.

shotType
string可选"customize"

Shot type.When the multiShot parameter is true, the current parameter is required.

可选值: "customize", "intelligence"

multiPrompt
object[]可选—

List of multi-prompt segments (optional).

≤ 6 items

elementList
object[]可选—

Element reference list (optional).

≤ 3 items

multiShot
boolean可选false

Whether to generate multi-shot video (when true, the prompt parameter is invalid; when false, the shotType parameter and the multiPrompt parameter are invalid)

Headers, response, task statuses and errors
请求头值说明
AuthorizationBearer <API key>也可以通过 x-api-key 发送密钥。
Content-Typeapplication/json
Idempotency-Key<unique string ≤128>必填。仅在重试完全相同的请求时复用。
字段类型说明
idstringHermes AI 任务 ID,用于轮询。
modelstring模型 slug。
statusstring参见任务状态。
errorCodestring | null任务失败或需要核对时设置。
priceMicrosinteger接收时锁定的报价,美元微单位。
currency"USD"
createdAt / updatedAtstringISO 8601 时间戳。
assets[]array{ id, type, url?, text? }。url 为同源 /api/v1/results/<uuid> 链接。
replayedboolean仅提交时返回。幂等重试命中时为 true。
状态终态含义
queued否已接收,等待执行。
submitting否正在提交给模型提供方。
running否模型生成中。
succeeded是已完成,结果位于 assets。
failed是未产生结果,预留额度已释放。
reconciliation_required是结果正在核实。请保留任务 ID,不要自动重新提交。
HTTP代码含义与处理方式
401AUTH_REQUIRED缺少或无效的 API Key 或会话。
403INSUFFICIENT_SCOPEAPI Key 缺少此端点所需的权限。
403PERMISSION_DENIED此凭据无权执行该操作。
429API_KEY_RATE_LIMITED已达到 API Key 的频率限制,请遵循 Retry-After。
429API_KEY_USAGE_EXCEEDEDAPI Key 已达到使用上限,请创建新密钥。
400INVALID_REQUEST请求体不是 JSON 对象,或缺少必需值。
400INVALID_IDEMPOTENCY_KEYIdempotency-Key 必须是 1–128 个可打印 ASCII 字符。
400INVALID_MODEL_INPUTinput 与模型 schema 不匹配(未知字段、缺少必填字段或值无效)。
400 / 413INPUT_TOO_LARGE请求体超过 256 KB。请先上传媒体并发送媒体引用。
404REQUEST_NOT_FOUNDNo chat request with this ID belongs to your account.
404MODEL_NOT_FOUND没有此 slug 对应的模型。
404JOB_NOT_FOUND你的账户中没有此 ID 的任务。
409IDEMPOTENCY_CONFLICT此 Idempotency-Key 已用于不同的输入。新请求请使用新密钥。
409PRICE_CHANGED价格与 expectedPriceMicros 不一致。请重新获取价格并确认。
409MODEL_PRICE_UNAVAILABLE此模型尚未配置价格,暂不可运行。
409PAYMENT_REVIEW_REQUIRED退款后余额不足。请联系支持或补足余额后再提交任务。
409INSUFFICIENT_CREDITS可用额度不足以支付此任务。
429RATE_LIMITED每分钟提交次数过多,请稍后重试。
429EXPORT_RATE_LIMITED本分钟内导出 CSV 次数过多,请按 Retry-After 等待后重试。
429CONCURRENCY_LIMITED未完成任务过多,请等待任务完成。
503EXECUTION_UNAVAILABLE此模型当前未配置执行能力。
429UPSTREAM_RATE_LIMITEDThe model provider is rate limiting chat requests. Retry with backoff; nothing was charged.
502UPSTREAM_ERRORThe model provider failed the chat request. Nothing was charged; retry later.
503PLATFORM_BUDGET_EXCEEDED此模型暂时不可用,请稍后重试。
503EXECUTION_TEMPORARILY_UNAVAILABLE执行暂时不可用,请使用相同的 Idempotency-Key 重试。
503AUTH_UNAVAILABLE身份验证暂时不可用,请稍后重试。
503API_UNAVAILABLEAPI 暂时不可用,请稍后重试。

Kling O3 API 常见问题

Answers built from the model's schema and the Hermes AI API contract.

Does every O3 variant have the same element controls?

No. The listed Standard, Pro and 4K schemas differ. Check the selected variant's parameter table before sending element references.

What can the Kling O3 API do?

Kling 的 Kling O3 支持文生视频, 图生视频, 参考生视频 和 视频编辑。目录列有11 variants, each,每个变体都有自己的模型 ID 和输入参数。提交请求前请先确认当前可用状态。

Which durations, resolutions and aspect ratios does Kling O3 support?

Across its variants, Kling O3 supports durations from 3 to 15 seconds 和 aspect ratios 1:1, 16:9 和 9:16. Each variant accepts its own subset; the variants table on this page lists them.

What input does Kling O3 accept?

A text prompt of up to 2,500 characters, plus 图像 或 视频 input (up to 7 images per request). Media fields take a public HTTPS URL or a media:<uuid> reference returned by POST /api/v1/media, so local files can be uploaded first. 未知字段会被拒绝并返回 INVALID_MODEL_INPUT。

How is Kling O3 priced and billed?

You pay per successful task from prepaid credits, with no subscription. The price depends on the variant and the specification you choose, such as resolution or duration; see the pricing table on this page or the pricing page. Credits are reserved when a task is accepted and charged only when it succeeds; a failed task returns the reservation to your balance.

Which Kling O3 variant should I choose?

For 文生视频, Standard text to video, Pro text to video 和 4K text to video differ in fields: Pro text to video adds elementList; 4K text to video adds elementList; for 图生视频, Standard image to video, Pro image to video 和 4K image to video differ in fields: Pro image to video adds elementList; 4K image to video adds elementList; for reference to video, 4K reference to video, Pro reference to video 和 Standard reference to video differ in fields: 4K reference to video adds elementList; Pro reference to video adds videoUrl, elementList; Standard reference to video adds videoUrl; for video edit, Pro video edit 和 Standard video edit differ in fields: Pro video edit adds elementList. Pick the task first, then compare the tiers in the variants table.

How do I call the Kling O3 API from code?

Send POST https://api.hermes-ai.net/api/v1/jobs with the variant's model ID and an input object, then poll GET /api/v1/jobs/{id} until the status is succeeded or failed. The API section has cURL, JavaScript and Python examples generated from the schema.

Can I try Kling O3 without writing code?

Yes. Sign in and use the playground on this page: it sends the same request as the API with your session and charges the same price. Code needs an API key from the dashboard.

What happens when a task fails or I hit a rate limit?

A failed task is not charged and its reserved credits are returned. Too many submissions per minute return 429 RATE_LIMITED, and too many unfinished tasks return 429 CONCURRENCY_LIMITED; retry later or wait for a task to finish.