Leina 
Wanta 
服务用量按单位从点数中扣除。
价格参考，更新于 2026年8月11日。实际扣费以用量明细为准。
点数如何消耗
点数用于内置媒体与语音服务，也用于超出月度额度的 App 调用。
| 服务名 | 价格（点数） | 说明 |
|---|---|---|
 | 0.0012 / 秒 | Upscale images with Aura SR, billed by inference time. |
 | CosyVoice 3.5 Flash: 0.144 / 万字 | Generate multilingual speech with CosyVoice models. |
 | 0.043 / 张图片 | Generate images from text prompts with Doubao Seedream. |
| | 0.34 / 小时 | Transcribe speech audio into text. |
| | 0.77 / 万字 | Generate speech audio from text with Doubao voices. |
 | 0.075 / 张图片 | Generate illustrations for text-to-EPUB workflows. |
| | 0.048 / 张图片 | Edit images with Flux Pro Kontext. |
 | Low: 0.012 / 张图片 | Generate and edit images with GPT Image 2. |
 | Read: 0.05 / M Token | Read web pages and search web content for agents. |
| | 9.9 / 千页 | Translate and optionally colorize manga pages from a ZIP archive. |
 | Nano Banana: 0.047 / 张图片 | Generate and edit images with Nano Banana models. |
| | 0.3 / 张图片 | Recognize text in images and optionally correct detected text. |
| | Next Generation: 0.144 / 万字 | Generate speech from text with OOMOL voice services. |
| | Generate: 0.04 / 张图片 | Generate or edit images through the OpenAI-compatible image API. |
 | ~2.5 / 千页 | Convert PDF documents into EPUB files. |
| | ~2.5 / 千页 | Convert PDF documents into Markdown files. |
| | Voice Cloning: 0.0018 / 个音色 | Create custom voices by cloning or describing a voice. |
| | Input: 0.1 / M Token | Extract and analyze information from documents with Qwen. |
| | Qwen Image 2.0: 0.034 / 张图片 | Generate images with Qwen Image models. |
| | Qwen Image Edit Plus: 0.034 / 张图片 | Edit images with Qwen image models. |
| | 0.06 / 张图片 | Separate an image into multiple visual layers. |
| | 0.0005 / 张图片 | Translate text in images while preserving the visual layout. |
| | 0.000039 / 秒 | Transcribe audio files into text with Qwen ASR. |
| | 0.144 / 万字 | Generate speech audio from text with Qwen voices. |
| | 0.021 / 张图片 | Cut out subjects and remove image backgrounds. |
 | Fast Image to Video: 3.96 / M Token | Generate videos from prompts and reference media with Seedance 2.0. |
| | Text to Video 720p: 0.36 / 秒 | Generate video clips from text prompts or a source image. |
| | 免费 | Store temporary files for free. Files may be removed periodically; each file can be up to 500 MB. |
 | 0.02 / 次请求 | Access supported public social data through TikHub. |
 | 0.009 / 张图片 | Compress PNG, JPEG, and WebP images. |
| | 0.035 / 张图片 | Generate and edit images with Wan models. |
| | Keyframe Wan 2.2 Flash 480p: 0.017 / 秒 | Generate videos from text, images, keyframes, or reference material with Wan models. |
App 调用是授权后使用已连接 App 的请求，包括 OAuth 回调、token 刷新和连接状态检查。
| 套餐 | 包含额度 | 超额价格 |
|---|---|---|
| 免费版 | 20K 次 / 月 | 0.3 / 1K 次 |
| Pro | 20K 次 / 月 | 0.3 / 1K 次 |
| Max | 20K 次 / 月 | 0.3 / 1K 次 |