Search FoxAI
AI Models 394
- wan3.0-video↗
阿里巴巴 · Qwen
万相3.0是 all-in-one 视频生成模型，统一支持参考、编辑、复刻、驱动等多种创作功能，四模态全能参考，最长支持30秒视频生成，可解析文件、网页及复杂图片。生产级角色一致性保持，音画真实，带来身临其境的视听冲击力。
- seedream-v5-pro/layer-decomposition↗
字节跳动
Seedream 5.0 pro 图层拆分：输入单张图片，拆分为1张底图+最多16个图层
- DeepSeek V4.1 Flash 1017 (Max)↗
DeepSeek
deepseek-v4.1-flash-1017 is the October 17 snapshot version of DeepSeek V4.1 Flash, a MoE multimodal model that supports text-and-image input, tool invocation, and million-token-long context. It comes with a built-in thinking mode and is designed for high throughput and cost-effectiveness, making it ideal for agent tasks, coding, long documents, and image understanding scenarios.
- gemini-3.1-flash-image-preview-2k↗
Google
Google最新的 nano-banana-2 第二代绘图模型，一致性极强, 快速版本，核心能力一览： 分辨率：最高 4K（4096×4096），与 Pro 持平参考图一致性：最多 14 张参考图（10 物体 + 4 角色），保持风格/角色一致极端比例：新增 1:4、4:1、1:8、8:1，适合长图、海报、Banner文字渲染：进阶文字生成，适合信息图、营销海报排版搜索增强：集成 Google Search + Image Search grounding思考链：内置 thinking process，复杂 prompt 先推理再生成 支持gemini、openai chat、openai dall-e 接口格式
- kling-image-v3-omni↗
快手 · 可灵
Kuaishou Kaling provides an officially authorized API that is secure and stable. It supports synchronous updates for all official models, and its billing methods are consistent with those of the official platform. To integrate, simply replace the official API domain and token with your own on-premises instance.
- claude-fable-5-1-thinking↗
Anthropic
Claude Fable 5.1 在 Claude Fable 5 的基础上进行了扩展，输入和输出价格保持不变，缓存读取成本降至四分之一，并带来了更强的长时间运行的智能体编码、多步骤研究，以及文档、电子表格和幻灯片处理能力。对于大多数工作负载
- gemini-nano-banana-2.1-4k↗
Google
The updated version of Nano Banana 2 is a leading, highly efficient model for image generation and conversational intelligent photo editing. It maintains Flash‑level speed and cost‑effectiveness at 1K, 2K, and 4K resolutions, while enhancing visual quality, text rendering, multi‑turn dialogue consistency, and Google Search relevance. Up to 10 high‑fidelity object images can be added to the final output, and up to 4 character images can be used to ensure consistent artistic style. It supports Gemini, OpenAI Chat, and OpenAI DALL‑E interface formats.
- gpt-image-2.5-sunburst-4k↗
OpenAI
GPT Image 2.5 Sunburst 它非常适合那些对编辑精度要求极高的应用场景 GPT Image 2.5 Flare 是我们目前用于生成高质量图片的最快模型 模型可以接受文本和图片作为输入，然后输出相应的图片
- MagiHuman↗
DeepBeepMeep
You will find here all the MagiHuman Video models used with WanGP (https://github.com/deepbeepmeep/Wan2GP) : WanGP supports the Wan (and derived models), Hunyuan Video, Flux 1 & 2, Qwen, Z-Image and LTV Video models with: Low VRAM requirements (as low as 6 GB of VRAM is sufficient for certain models) Support for old GPUs (RTX 10XX, 20xx, ...) Very Fast on the latest GPUs Easy to use Full Web based interface Auto download of the required model adapted to your specific architecture Tools integrated to facilitate Video Generation : Mask Editor, Prompt Enhancer, Temporal and Spatial Generation Loras Support to customize each model Queuing system : make your shopping list of videos to generate and come back later Discord Server to get Help from Other Users and show your Best Videos: https://discord.gg/g7efUW9jGV
- CyberRealisticXL↗
cyberdelia
**CyberRealistic XL** is a photorealistic image generation model based on **Stable Diffusion XL 1.0 (SDXL)**. It is designed to produce highly realistic, cinematic-quality outputs with minimal prompt engineering. This checkpoint is optimized for portraits, fashion, and editorial-style scenes, but also handles a wide range of creative prompts with remarkable fidelity. - **Model Type**: Image generation (Text-to-Image) - **Base Model**: Stable Diffusion XL 1.0 - **Format**: `safetensors` - **Latest Version**: v5.7 - **Creator**: Cyberdelia - **File Size**: ~6.31 GB - **License**: CreativeML Open RAIL++-M License - ⚡ **Ultra-Photorealism** – Exceptional detail in faces, lighting, and textures. - 🧩 **Effortless Prompting** – Works well with short prompts, but supports complex ones too. - 🪞 **Skin & Eye Fidelity** – Improved clarity in eyes, skin detail, and facial structure. - 🔧 **Built-in VAE** – No need for an external VAE; optimized out of the box.