grok-imagine-video AI 视频 API
grok-imagine-video 可通过 Flatkey 使用,适用于视频生成;Flatkey 通过 /v1/videos 路由此模型,供应商为 xAI。$0.072 per second。
输出
预览使用 Seedance 2.0 生成
请求摘要
- 接口
- POST /v1/videos
- 模型 ID
- grok-imagine-video
- 画面比例
- 16:9
- 分辨率
- —
- 时长
- 5s
- 参考媒体
- 无
仅供预览。登录后即可向 Flatkey 提交此请求。
{
"model": "grok-imagine-video",
"prompt": "Creative direction: A man in a dark gallery pushes and rotates a glowing turquoise rectangular panel. Orange and purple panels light up around him as he changes its angle, ending with the three colored surfaces arranged across the room.\n\nVisual treatment: Light-installation performance with turquoise, orange and purple emissive panels.\n\nShot sequence: 1. Frame the man touching the turquoise panel in the dark gallery.; 2. Follow his hands pushing and rotating the panel while orange and purple panels illuminate.; 3. Hold all three colored surfaces in their final arrangement.\n\nContinuity: Keep the main subject, palette and light behavior consistent as the visual pattern develops.\nExclude: extra people or props, text, logos and watermarks.",
"duration": 5
}API 性能
grok-imagine-video API 可靠性与可用性
当生产流量达到足够规模后,这里会显示 grok-imagine-video 的 Flatkey 实时请求遥测数据。
API 活动
grok-imagine-video API 使用量与请求活动
查看最近统计窗口内 grok-imagine-video 的请求量和成功推理活动。
grok-imagine-video 价格
grok-imagine-video API 价格与计费
下方价格来自当前模型的 Flatkey 定价数据,以及定价 API 返回的可见分组。
充值余额
图像、视频、音频和文本模型共用同一个 Flatkey 余额和 API Key。
- 模型类型
- 文生视频
- API
- /v1/videos
- 计费依据
- 秒
- 时长
- 1–15s
- 画面比例
- adaptive
grok-imagine-video AI 视频生成器与 API: 核心能力
xAI 的 grok-imagine-video 路由支持 文本 · 图像 · 视频。目录分类:Marketing。
grok-imagine-video视频生成
通过 /v1/videos 将文字简报转换为视频生成结果。
grok-imagine-video参考工作流
当 grok-imagine-video 路由接受参考媒体或上下文时,可使用已有参考进行创作。
grok-imagine-video渠道适配版本
将一个创意调整为店铺、营销活动或内容流水线所需的不同版本。
grok-imagine-video生产交接
将确认后的结果从提示词探索带入可重复的 Flatkey 生成工作流。
对比
grok-imagine-video API 对比:能力与访问方式
在调整集成前,将 grok-imagine-video 的目录事实与上一代基线进行对比。
| 能力 | 上一代 | grok-imagine-video |
|---|---|---|
| 供应商 | 此目录快照未验证 | xAI |
| 模态 | 此目录快照未验证 | 文本 · 图像 · 视频 |
| 上下文 | 此目录快照未验证 | 目录未列出上下文窗口 |
| 端点 | 此目录快照未验证 | /v1/videos |
| 计费 | 此目录快照未验证 | $0.072 per second |
接入前先选对请求方式
核对支持的工作流、当前配置的实时费用,以及 Flatkey 实际使用的参数契约。
当目录提供固定单价时,根据当前 Playground 配置和实时目录价格计算。
- 当前配置
- 5s
- 实时单价
- $0.072 / 秒
- 本次预估
- $0.360 $0.072 × 5s
最终费用以实际提交的请求和账户分组为准。Token 或分层计费无法简化成一个固定预估值。
工作流与 Model ID
只使用标记为可用的工作流。Flatkey 对外保持一个模型 ID,由请求体选择输入方式。
文生视频
通过文本提示词和支持的输出参数生成视频。
- 模型 ID
grok-imagine-video- 接口
/v1/videos
图生视频
让一张或多张参考图动起来,并尽量保持主体与构图。
- 模型 ID
grok-imagine-video- 接口
/v1/videos
参数兼容矩阵
以下取值与 Playground 和请求预览使用同一份模型配置。
duration时长1–15
默认值: 5image参考素材上限1 个文件
接口: /v1/videos从首次测试到生产接入
Playground 与 API 使用同一套请求契约,并明确处理参数校验、余额、任务状态和产物获取。
请求生命周期
- 1
校验模型、提示词、参考素材和输出设置。
- 2
提交 POST 请求并保存返回的任务 ID。
- 3
轮询任务状态或处理已配置的回调。
- 4
从产物接口获取文件,并在失效前保存到自己的存储。
生产检查项
提交前拦截不支持的尺寸、时长、比例、格式或参考素材数量。
分别处理无效密钥、余额不足、限流和账户分组价格。
明确指出被拒绝的提示词或素材,让用户只替换失败的输入。
保留任务 ID,限制重试次数,且不能把超时当成生成成功。
提示词库
grok-imagine-video 视频生成提示词示例
先使用一个面向视频生成的 grok-imagine-video 提示词,再调整 Playground 中显示的请求字段。
使用 Seedance 2.0 生成
Creative direction: A young man in a mustard jacket stands in an empty stairwell holding a white envelope. He opens it, studies the contents, and looks upward toward the landing as the camera moves from a medium view into his reaction. Visual treatment: Live-action suspense with subdued stairwell light and a mustard jacket accent. Shot sequence: 1. Frame the mustard-jacketed man and white envelope in the stairwell.; 2. Ease closer as he opens the envelope and reads.; 3. Hold his upward glance toward the landing without changing eyeline direction. Continuity: Keep the character's face, clothing and key prop consistent; preserve screen direction across cuts. Exclude: extra people or props, text, logos and watermarks.
使用 Seedance 2.0 生成
Creative direction: A white travel mug with a blue hinged lid stands on a dark studio surface. One hand flips the lid open at its fixed hinge; a thin plume of steam rises from the opening. Finish on the mug profile with its lid, handle and opening clearly visible. Visual treatment: Photoreal product close-up with a soft edge light on white ceramic, a blue lid and restrained backlit steam. Shot sequence: 1. Hold the closed mug in a three-quarter close-up; 2. Show the hand opening the lid around its hinge; 3. Let the steam rise and settle on the open product profile Continuity: Keep the product silhouette, cap or hinge, materials and prop positions stable; hold the final hero frame. Exclude: extra people or props, text, logos and watermarks.
使用 Seedance 2.0 生成
Creative direction: A man enters an empty industrial room beside the sea and walks toward a large window. The camera follows from behind as the ocean and warm sunset fill the opening, ending on his silhouette against the view. Visual treatment: Live-action location reveal with industrial shadows and warm coastal sunset. Shot sequence: 1. Follow the man from behind through the industrial room.; 2. Approach the large window as the sunset ocean fills the opening.; 3. Stop behind his silhouette and retain detail around the window frame. Continuity: Maintain scene geography, subject scale and motivated light direction throughout the camera move. Exclude: extra people or props, text, logos and watermarks.
使用 Seedance 2.0 生成
Creative direction: A man stands still in a neutral gray studio while a compact quadcopter drone circles behind and beside him. Keep the camera locked at waist height so the drone's orbit, distance, and the subject's profile remain easy to read. Visual treatment: Live-action character-and-drone study with neutral gray lighting and clear spatial separation. Shot sequence: 1. Hold the man in profile in the neutral gray studio.; 2. Move the quadcopter behind and beside him along a clear curved path.; 3. Keep its distance from the man as the locked camera holds the composition. Continuity: Maintain character design, equipment geometry and travel direction from first to final frame. Exclude: extra people or props, text, logos and watermarks.
使用 Seedance 2.0 生成
Creative direction: A presenter speaks at a home desk into a black microphone. Cut to a close view of one hand connecting a small audio cable and adapter, then return to the seated presenter continuing the explanation. Visual treatment: Live-action creator demonstration with natural desk lighting and restrained depth of field. Shot sequence: 1. Establish the seated presenter speaking into the black microphone.; 2. Cut to the hand aligning and inserting the cable into the adapter.; 3. Return to the original desk framing as the explanation continues. Continuity: Keep demonstration props, layout and hand contact consistent so the action remains legible. Exclude: extra people or props, text, logos and watermarks.
使用 Seedance 2.0 生成
Creative direction: A man in a dark gallery pushes and rotates a glowing turquoise rectangular panel. Orange and purple panels light up around him as he changes its angle, ending with the three colored surfaces arranged across the room. Visual treatment: Light-installation performance with turquoise, orange and purple emissive panels. Shot sequence: 1. Frame the man touching the turquoise panel in the dark gallery.; 2. Follow his hands pushing and rotating the panel while orange and purple panels illuminate.; 3. Hold all three colored surfaces in their final arrangement. Continuity: Keep the main subject, palette and light behavior consistent as the visual pattern develops. Exclude: extra people or props, text, logos and watermarks.
为什么使用 grok-imagine-video API
为什么通过 Flatkey 使用 grok-imagine-video?
通过一个网关管理 grok-imagine-video、账户控制以及模型目录中的其他模型。
一个 API 使用 grok-imagine-video
明确保留 grok-imagine-video 模型 ID 和端点,同时与其他工作负载共用 Flatkey 密钥。
实时目录价格
发送请求前查看当前 grok-imagine-video 计费维度,并在账户中确认最终预估。
实用的模型交接
在公开 Playground 中测试 grok-imagine-video 提示词,再将相同设置带入经过身份验证的集成。
用量与路由控制
集中管理 grok-imagine-video 的密钥、配额和路由,无需改变应用面向供应商的工作流。
grok-imagine-video API
grok-imagine-video API 集成指南
使用上方模型 ID 调用 /v1/videos;为当前路由保留已记录的请求字段和计费单位。
curl https://router.flatkey.ai/v1/videos \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-***" \ -d '{"model":"grok-imagine-video","prompt":"Creative direction: A young man in a mustard jacket stands in an empty stairwell holding a white envelope. He opens it, studies the contents, and looks upward toward the landing as the camera moves from a medium view into his reaction.\n\nVisual treatment: Live-action suspense with subdued stairwell light and a mustard jacket accent.\n\nShot sequence: 1. Frame the mustard-jacketed man and white envelope in the stairwell.; 2. Ease closer as he opens the envelope and reads.; 3. Hold his upward glance toward the landing without changing eyeline direction.\n\nContinuity: Keep the character'"'"'s face, clothing and key prop consistent; preserve screen direction across cuts.\nExclude: extra people or props, text, logos and watermarks.","duration":5}' # Poll for the result (replace TASK_ID with the returned task id) curl "https://router.flatkey.ai/v1/videos/TASK_ID" \ -H "Authorization: Bearer sk-***"
常见问题
grok-imagine-video API
价格、功能与使用常见问题
关于 grok-imagine-video 价格、能力、端点访问和目录限制的答案。