实时 AI 模型目录
图像模型
28
Nano Banana Pro
文本生成图像Generate images from text prompts using Nano Banana Pro.Nano Banana 2
文本生成图像Generate images from text prompts using Nano Banana 2.
Qwen2 Text-to-Image
文本生成图像Generate images from text prompts using Qwen2.Qwen2 Image Edit
图像转换图像Edit images using text prompts with Qwen2.
Seedream 5.0 Lite
文本生成图像Generate high-quality photorealistic images from text prompts using Seedream 5.0 Lite.Seedream 5.0 Lite Edit
图像转换图像Generate images from reference images using Seedream 5.0 Lite.
Flux 2 Pro Text-to-Image
文本生成图像Generate photorealistic images from text prompts using Flux 2 Pro.Flux 2 Pro Image-to-Image
图像转换图像Generate images from reference images using Flux 2 Pro.
Grok Imagine Image 2.0
文本生成图像Generate images with xAI Grok Imagine Image 2.0 at 1K or 2K quality.Grok Imagine Text-to-Image
文本生成图像Generate images from text prompts using xAI Grok Imagine.Grok Imagine Image 2.0 Edit
图像转换图像Edit images with up to three references using Grok Imagine Image 2.0.Grok Imagine Image-to-Image
图像转换图像Generate images from reference images using xAI Grok Imagine.
GPT Image 1.5 Text-to-Image
文本生成图像Generate images from text prompts using GPT Image 1.5.GPT Image 1.5 Image-to-Image
图像转换图像Generate images from reference images using GPT Image 1.5.GPT Image 2 Text-to-Image
文本生成图像Generate images from text prompts using GPT Image 2.GPT Image 2 Image-to-Image
图像转换图像Generate images from reference images using GPT Image 2.
Wan 2.7 Image
文本生成图像Generate or edit images using Wan 2.7. Connect or upload reference images for editing.Wan 2.7 Image Pro
文本生成图像Higher-quality image generation and editing with Wan 2.7 Pro. Adds sequential batch mode, thinking mode, and 8:1/1:8 ultrawide aspect ratios.Flux 2 Flex Text-to-Image
文本生成图像Generate photorealistic images from text prompts using Flux 2 Flex.Flux 2 Flex Image-to-Image
图像转换图像Generate images from reference images using Flux 2 Flex.GPT Image 2
文本生成图像Generate images from text prompts using GPT Image 2 (Runware).GPT Image 2
图像转换图像Generate images from reference images using GPT Image 2 (Runware).Seedream 5.0 Pro
文本生成图像Generate high-quality photorealistic images from text prompts using Seedream 5.0 Pro.Nano Banana 2 Lite
文本生成图像Fast, lower-cost image generation with Nano Banana 2 Lite.Seedream 5.0 Pro Edit
图像转换图像Generate images from reference images using Seedream 5.0 Pro.
Ideogram V3 Remix
图像转换图像Remix a source image with Ideogram V3 using an exact rendering-speed tier.Qwen3 Pro Text to Image
文本生成图像Generate images from text with Qwen3 Pro at 1K or 2K resolution.Qwen3 Pro Image to Image
图像转换图像Edit up to three source images with Qwen3 Pro at 1K or 2K resolution.视频模型
54Seedance 2.5 Text-to-Video
文本生成视频Generate videos from text prompts using ByteDance Seedance 2.5.Seedance 2.5 Image-to-Video
图像生成视频Generate videos from first/last frames or multimodal image and audio references using ByteDance Seedance 2.5.
Kling 3.0 Omni Transformation
视频转换视频Transform a source video with text guidance using Kling 3.0 Omni.Seedance 2.5 Video-to-Video
视频转换视频Generate videos from video, image, and audio references using ByteDance Seedance 2.5.Kling 3.0 Omni Text-to-Video
文本生成视频Generate videos from text prompts using Kling 3.0 Omni.Kling 3.0 Omni Image-to-Video
图像生成视频Animate a first frame or first/last frame pair using Kling 3.0 Omni.Kling 3.0 Omni Reference-to-Video
图像生成视频Generate a consistent video from reference images or video using Kling 3.0 Omni.Kling 3.0
文本生成视频Generate videos using Kling 3.0 with standard or pro quality.Seedance 1.5 Pro
文本生成视频Generate videos with audio using Seedance 1.5 Pro by ByteDance.Kling 3.0 Motion Control
视频转换视频Drive a subject image with the motion of a source video using Kling 3.0.Wan 2.6 Text-to-Video
文本生成视频Generate videos from text prompts using Wan 2.6.Kling 3 Turbo Text-to-Video
文本生成视频Generate videos from text prompts using Kling 3 Turbo.Wan 2.6 Image-to-Video
图像生成视频Generate videos from images using Wan 2.6.Kling 3 Turbo Image-to-Video
图像生成视频Generate videos from images using Kling 3 Turbo.Wan 2.6 Video-to-Video
视频转换视频Generate videos from source videos using Wan 2.6.Kling 2.6 Text-to-Video
文本生成视频Generate videos from text prompts using Kling 2.6.Kling 2.6 Image-to-Video
图像生成视频Generate videos from images using Kling 2.6.Veo 3.1
文本生成视频Generate high-fidelity videos with Google Veo 3.1 from a text prompt, reference images, or a first + last frame.Grok Imagine Text-to-Video
文本生成视频Generate videos from text prompts using xAI Grok Imagine.Grok Imagine Image-to-Video
图像生成视频Generate videos from images using xAI Grok Imagine.Seedance 2.0 Text-to-Video
文本生成视频Generate videos from text prompts using Bytedance Seedance 2.0.Seedance 2.0 Image-to-Video
图像生成视频Generate videos from images using Bytedance Seedance 2.0.Seedance 2.0 Video-to-Video
视频转换视频Generate videos from source videos using Bytedance Seedance 2.0.Wan 3.0
文本生成视频Generate videos from text, frames, references, files, or links using Wan 3.0 in Normal or Speed mode.Seedance 2.0 Fast Text-to-Video
文本生成视频Generate videos from text prompts using Bytedance Seedance 2.0 Fast (lower-cost, faster variant).Wan 2.7 Text-to-Video
文本生成视频Generate videos from text prompts using Wan 2.7.Seedance 2.0 Fast Image-to-Video
图像生成视频Generate videos from images using Bytedance Seedance 2.0 Fast (lower-cost, faster variant).Wan 2.7 Image-to-Video
图像生成视频Generate videos from images using Wan 2.7.Seedance 2.0 Fast Video-to-Video
视频转换视频Generate videos from source videos using Bytedance Seedance 2.0 Fast (lower-cost, faster variant).Seedance 2.0 Multimodal Reference
图像生成视频Generate videos from reference images, audio, and optional video clips using Bytedance Seedance 2.0.HappyHorse 1.0 Text-to-Video
文本生成视频Generate videos from text prompts using HappyHorse 1.0 (Alibaba) via kie.ai.Seedance 2.0 Fast Multimodal Reference
图像生成视频Generate videos from reference images, audio, and optional video clips using Bytedance Seedance 2.0 Fast (lower-cost, faster variant).HappyHorse 1.0 Image-to-Video
图像生成视频Generate videos from images using HappyHorse 1.0 (Alibaba) via kie.ai.Seedance 2.0 Mini Text-to-Video
文本生成视频Generate videos from text prompts using Bytedance Seedance 2.0 Mini (fast, cost-efficient variant).Seedance 2.0 Mini Image-to-Video
图像生成视频Generate videos from images using Bytedance Seedance 2.0 Mini (fast, cost-efficient variant).Gemini Omni
文本生成视频Generate multimodal videos from text + reference images using Gemini Omni.Wan 2.7 Video Edit
视频转换视频Edit source videos with text guidance using Wan 2.7.
Topaz Video Upscale
视频转换视频Upscale videos using Topaz Labs AI via kie.ai.Seedance 2.0 Mini Video-to-Video
视频转换视频Generate videos from source videos using Bytedance Seedance 2.0 Mini (fast, cost-efficient variant).Grok Imagine Video 1.5 Image-to-Video
图像生成视频Animate a source image at up to native 1080p with Grok Imagine Video 1.5.Wan 2.7 Reference-to-Video
图像生成视频Generate videos from reference images/videos using Wan 2.7.HappyHorse 1.0 Reference-to-Video
图像生成视频Generate videos from up to 9 reference images using HappyHorse 1.0 (Alibaba) via kie.ai.HappyHorse 1.0 Video Edit
视频转换视频Edit source videos with HappyHorse 1.0 (Alibaba) via kie.ai.Seedance 2.0 Mini Multimodal Reference
图像生成视频Generate videos from reference images, audio, and optional video clips using Bytedance Seedance 2.0 Mini (fast, cost-efficient variant).Grok Imagine Video 1.5 Reference-to-Video
图像生成视频Generate a video from up to seven reference images with Grok Imagine Video 1.5.HappyHorse 1.1 Text-to-Video
文本生成视频Generate videos from text prompts using HappyHorse 1.1 (Alibaba) via kie.ai.HappyHorse 1.1 Image-to-Video
图像生成视频Generate videos from images using HappyHorse 1.1 (Alibaba) via kie.ai.HappyHorse 1.1 Reference-to-Video
图像生成视频Generate videos from up to 9 reference images using HappyHorse 1.1 (Alibaba) via kie.ai.
PixVerse V6 Text-to-Video
文本生成视频Generate videos from text prompts using PixVerse V6.PixVerse V6 Image-to-Video
图像生成视频Generate videos from images or templates using PixVerse V6.PixVerse V6 Transition
图像生成视频Generate a transition between first and last frame images using PixVerse V6.PixVerse V6 Extend
视频转换视频Extend an uploaded video or a completed PixVerse task using PixVerse V6.PixVerse V6 Reference-to-Video
图像生成视频Generate videos from ordered image references using PixVerse V6.Gemini Omni 1.1 Flash
文本生成视频Character-free video generation. Character IDs are rejected. With reference video, output duration is determined by the model, not the requested duration.音频模型
1
文本模型
14GPT-5.6 Sol
文本生成Flagship GPT-5.6 model. Frontier reasoning and multimodal intelligence for the hardest coding and professional work. 1M input / 128K output, vision.GPT-5.6 Terra
文本生成Balanced GPT-5.6 model: strong reasoning at lower cost for everyday professional workloads. 1M input / 128K output, vision.GPT-5.6 Luna
文本生成Efficient GPT-5.6 model for fast, high-volume tasks. 1M input / 128K output, vision.GPT-5.5
文本生成Frontier multimodal model. A new class of intelligence for coding and professional work. 1M input / 128K output, vision.GPT-5.4
文本生成Frontier model for complex professional work, balancing advanced reasoning with affordability. 1.05M input / 128K output, vision.GPT-5.4 mini
文本生成Strongest mini model for coding, computer use, and subagents. 272K input / 128K output, vision.GPT-5.4 nano
文本生成Cheapest GPT-5.4-class model for simple high-volume tasks. 272K input / 128K output, vision.Grok 4.6
文本生成xAI's flagship reasoning model with a 500K-token context window, text and image input, and configurable reasoning through extra-high effort.Grok 4.5
文本生成xAI's newest flagship Grok model with a 500K-token context window and stronger reasoning. Multimodal (text + image input).Grok 4.3
文本生成xAI's most intelligent and fastest Grok model. Multimodal (text + image input).
Claude Opus 4.8
文本生成Anthropic's most capable Claude model. Best-in-class for complex coding, agentic workflows, and long-horizon reasoning. 1M context, vision.Claude Opus 4.7
文本生成A highly capable Claude model for complex coding, agentic workflows, and long-horizon reasoning. 1M context, vision.Claude Sonnet 4.6
文本生成Balanced Claude model for production workloads: strong reasoning, lower latency, large context. 1M context, vision.Claude Haiku 4.5
文本生成Fastest, cheapest Claude model for high-volume tasks. 200K context, vision.比较当前模型
使用同一 Flowctory 实时目录中的事实并排比较，不做主观质量排名。