Cookie Preferences
We use cookies to personalize content and analyze traffic. See our Cookie Policy.
还没有输出。提交表单以生成内容。
Gemini Omni 使用完整指南
将一张参考图和一段简短描述转化为可复用的 AI 角色，并可在 ApiPass 上接入每一次 Gemini Omni 视频生成。
ApiPass 上的 Gemini Omni Character 可让你通过一段简短描述和一张参考图创建可复用的角色资源。你无需在每次调用时重新描述角色或重新上传照片，只需注册一次角色，即可获得一个精简资源，并在未来任何 Gemini Omni Video 请求中重复使用——仍然使用你已用于 ApiPass 模型目录其余服务的同一个 API 密钥。
只需注册一次角色，即可在未来每次 Gemini Omni Video 调用中引用它——无需重复上传照片，也无需重复输入描述。
将角色绑定到参考图，让同一张脸、同一种风格和同一个身份在不同片段和场景之间自然延续。
角色资源可与通过 Gemini Omni Audio 创建的语音档案自然配合，让角色在整个系列或营销活动中同时保持一致的外观和声音。
每个创建的角色都会返回一个精简资源，便于持久化保存到你自己的数据库，并展示在产品 UI 中。
角色创建与 Gemini Omni Video 和 Gemini Omni Audio 位于同一个 ApiPass 接入界面中，因此你可以构建端到端的角色驱动流程，而无需在多个提供商之间切换。
使用一张清晰的参考图，加上一段涵盖外观、风格和性格的自然语言描述，注册一个新角色。
创建资源时为每个角色命名，便于在你自己的素材库或 UI 中识别。
将角色与 Gemini Omni Audio 的语音档案配对，让声音人格从一开始就与角色关联。
API 会返回一个可复用的角色资源，可附加到任何 Gemini Omni Video 请求中，按需驱动角色一致的生成结果。
多个已注册角色可组合到单个 Gemini Omni Video 请求中，让包含多个常驻角色的场景也能共同保持一致表现。
在 ApiPass 上创建一次你的 AI 虚拟形象，然后让它替你主持视频。无论你在运营 YouTube 频道、TikTok 系列，还是每周播客，你的角色都会在每一集中稳定出现——同一张脸、同一种声音、同一种气质。不用再架设相机，也不用担心发型状态不好。只需写好脚本，让你的 AI 分身来开口表达。
一个经过批准的统一角色，可出现在广告变体、产品教程和本地化市场内容中，并在所有出现位置保持品牌身份一致。
为系列化短视频内容和连载故事世界创建主角、搭档和主持人，并让他们在每一集之间始终保持设定一致。
为课程、讲解视频和培训模块创建一致的教师、旁白和主持人，让每一课都像由同一张熟悉面孔讲授。
构建系列化内容，让同一位主持人或主角在每个视频中都能稳定出现。
维护符合品牌调性的代言人或吉祥物，让其出演每支广告、讲解视频和本地化版本。
设计一致的讲师或演示者人格，用于贯穿整个课程目录。
注册会反复出现的角色，用于过场动画、预告片和应用内剧情序列。
为终端用户提供“我的角色”库，用于保存、管理和复用他们自己的 AI 身份。
创建 ApiPass 账户，并从仪表板生成 API 密钥，即可通过单一凭证解锁完整的 Gemini Omni 系列——角色、视频和音频。
选择一张清晰的参考图，编写一段涵盖外观和风格的简短描述，并选择一个角色名称。也可以选择将角色与 Gemini Omni Audio 的语音档案配对，打造更完整的身份。
调用 Gemini Omni Character 注册角色，然后将其附加到未来任何 Gemini Omni Video 请求中，以规模化生成一致的角色驱动视频。
所有 API 都需要通过 Bearer Token 进行身份验证。
Authorization: Bearer
Create a reusable Gemini Omni character resource from a character description and one public reference image. The returned characterId can be used as a character_ids value with google/gemini-omni-video.
1{
2 "model": "google/gemini-omni-character",
3 "channel": "auto",
4 "input": {
5 "descriptions": "A young female character with short silver hair, a black futuristic utility jacket, calm expression, agile posture, and a clean cyberpunk visual style.",
6 "image_urls": [
7 "https://cdn.apipass.dev/apipass/results/task_78514b0d12c44ad4_0.png"
8 ],
9 "character_name": "Jenny"
10 }
11}model必填stringModel endpoint name. Use google/gemini-omni-character for this endpoint.
google/gemini-omni-character
channel可选stringDefault channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics; starter is ultra-low-cost with limited quotas and weaker stability; regular is standard and much cheaper than official APIs with moderate stability; official uses the model native API with high stability, fast task execution, and official pricing.
Default channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics. starter is ultra-low-cost with limited quotas and weaker stability. regular is standard and much cheaper than official APIs with moderate stability. official uses the model native API with high stability, fast task execution, and official pricing.
可用选项：
auto
callBackUrl可选stringOptional callback URL for receiving task result updates. For this model, callback behavior depends on the adapter implementation.
https://example.com/callback
descriptions必填stringCharacter description used to define the appearance, identity, style, clothing, or personality of the character.
Use the plural field name descriptions. The official schema requires descriptions and image_urls; although some examples may show description, descriptions is the recommended field name.
A young female character with short silver hair and a futuristic utility jacket, calm, agile, and strongly cyberpunk in style.
image_urls必填array<string>Array of public character reference image URLs. Only 1 image is supported, and each image must be no larger than 20MB.
The image URL must be publicly accessible by the upstream service. Maximum items: 1.
[ "https://cdn.apipass.dev/apipass/results/task_78514b0d12c44ad4_0.png" ]
audio_ids可选array<string>Array of audio IDs generated by the google/gemini-omni-audio endpoint. These can provide voice traits, tone, or persona guidance for the character.
audio_ids must come from the corresponding KIE gemini-omni-audio resource.
[ "audio_xxx" ]
character_name可选stringCharacter name.
Jenny
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Authorization: Bearer YOUR_API_KEY" \
3 -H "Content-Type: application/json" \
4 -d '{
5 "model": "google/gemini-omni-character",
6 "channel": "auto",
7 "input": {
8 "descriptions": "A young female character with short silver hair, a black futuristic utility jacket, calm expression, agile posture, and a clean cyberpunk visual style.",
9 "image_urls": [
10 "https://cdn.apipass.dev/apipass/results/task_78514b0d12c44ad4_0.png"
11 ],
12 "character_name": "Jenny"
13 }
14}'1{
2 "code": 200,
3 "msg": "success",
4 "data": {
5 "characterId": "b09dbf56...",
6 "characterName": "Jenny",
7 "imageUrl": "https://xx.com/a.png"
8 }
9}codeResponse status code. 200 indicates a successful upstream character creation response.
msgResponse message.
dataCreated character resource information.
data.characterIdGenerated character ID. Use this value in google/gemini-omni-video input.character_ids when generating videos with this character.
data.characterNameCharacter name returned by the provider.
data.imageUrlURL of the character reference or generated character image returned by the provider.
ByteDance
ByteDance
起价
0 积分
ByteDance
ByteDance
起价
0 积分
ByteDance
ByteDance
Turn text, images, video clips, and audio into cinematic, audio-synced videos with a single API call.
起价
0 积分
MiniMax
MiniMax
起价
0 积分
ByteDance
ByteDance
Generate cinematic 4K AI videos up to 30 seconds with Seedance 2.5's multimodal power.
起价
0 积分
Google Veo 3.1 is the latest state-of-the-art generative video model designed to transform text and image prompts into high-fidelity cinematic visuals. Building upon its predecessors, Veo 3.1 features significant upgrades in prompt adherence, visual realism, and native audio generation, creating 8-second clips with synchronized sound effects and dialogue.
起价
0 积分
wan
wan
起价
0 积分
Runway
Runway
起价
0 积分
Kling
Kling
Transform text prompts and images into high-quality videos with advanced AI. Generate cinematic content with customizable duration, aspect ratio, and audio capabilities.
起价
0 积分
MiniMax
MiniMax
Generate high-quality AI videos from text prompts or images with MiniMax's Hailuo 2.3 model. Support for both 6s and 10s durations, 768p and 1080p resolutions, with intelligent prompt optimization.
起价
0 积分
grok
grok
Grok Imagine is xAI’s AI video generation model for creating high-quality videos with realistic motion, native audio, and synchronized speech. Grok Imagine 1.5 introduces faster generation speeds, improved motion physics, and stronger visual consistency for modern AI video applications.
起价
0 积分
Luma
Luma
起价
0 积分
Kling
Kling
Kling v3 video generation model, supports text-to-video and image-to-video, up to 15 seconds, 1080p pro mode
起价
0 积分
wan
wan
Generate videos with cinematic consistency by reimagining existing footage through high-fidelity visual transformation and structural control.
起价
0 积分
grok
grok
起价
0 积分
kling
kling
起价
0 积分