全部 AI 模型，開箱即用

增强视频分辨率、清晰度和细节，适合低清片段修复和高清交付。

Seedance 2.0 is ByteDance's cinematic audio-video generation lineup. It pairs stronger prompt comprehension with director-level camera moves, realistic physics, native soundtrack generation, and multi-shot continuity so every clip feels like a finished scene.

面向复杂视觉任务，突出世界知识、本地化、品牌一致性和精细创意控制。

通过 muyu-fal 调用 ai-route 的 Midjourney V7 生图模型，支持提示词和可选参考图。

OpenAI GPT-Image is OpenAI's image generation and editing lineup, designed for strong instruction following, high-quality visual output, natural-language image edits, and reliable text rendering inside images.

增强图像分辨率与纹理细节，适合低清素材修复和高清放大。

xAI Grok Imagine image models for text-to-image generation and reference-image editing.

豆包 Seedream 5.0 Pro 图片生成，通过 Fal-compatible 图片网关调用。

即梦图片

即梦 Seedream 4.6 图像模型，面向高质量视觉创意、精细编辑和风格化生成。

ElevenLabs TTS is ElevenLabs' text-to-speech lineup for natural voice synthesis with high intelligibility, expressive delivery, and timing-aware controls for production workflows.

MiniMax TTS is MiniMax's text-to-speech lineup for high-quality multilingual voice synthesis, covering both quality-first and speed-first generation workflows.

Qwen TTS is Qwen's text-to-speech lineup for controllable voice generation, supporting preset voices and speaker-clone style workflows with tunable decoding settings.