SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,708 results · Page 943 of 992

IF

AI images

DeepFloyd IF is an open-source text-to-image model launched by StabilityAI's DeepFloyd Lab, with high photorealism and language understanding capabilities. It consists of a frozen T5 text encoder and three cascaded pixel diffusion modules: the base model generates 64x64 pixel images based on text prompts, and two super-resolution modules upscale the images to 256x256 and 1024x1024 pixels, respectively. All stages use a frozen text encoder based on the T5 transformer to extract text embeddings, which are then fed into a UNet architecture enhanced with cross-attention and attention pooling. According to sources, the model achieves a zero-shot FID score of 6.66 on the COCO dataset. The minimum requirements for using all IF models are: IF-I-XL and IF-II-L require 16GB of VRAM, plus Stable x4 super-resolution to 1024x1024 requires 24GB of VRAM, and xformers must be installed and the environment variable FORCE_MEM_EFFICIENT_ATTN=1 must be set. The repository provides local Jupyter Notebook support for Dream, style transfer, super-resolution, and inpainting modes, and integrates with the Hugging Face Diffusers library, allowing stages to be run separately and intermediate results to be inspected. Model cards and weights will be released successively under the DeepFloyd organization on Hugging Face. The initial version uses a restricted license limited to research use only, with plans to release a fully open-source model later. The model has known limitations and biases, as detailed in the model card.

Read more

Imagable

Other AI tools

Imagable is a one-stop AI content creation platform that provides tools for image generation, video generation, image editing, and enhancement. The platform supports multiple input methods such as text-to-image, image-to-image conversion, text-to-video, and image animation, and integrates more than 15 mainstream models including Flux Pro, Ideogram V3, Sora 2, Veo 3.1, and Kling 2.5. Users can control aspect ratio, image quality, and generation parameters in a unified interface, and download in multiple formats once finished. Basic tools such as background removal, image upscaling, and image-to-prompt have daily free usage quotas, while advanced models and video generation consume credits, which can be obtained through subscriptions or one-time credit packs. Generated content can be used for commercial purposes, users own the copyright, the platform supports 720p, 1080p, and 4K video output, and provides enterprise-grade security protection.

Read more

Imagine

AI developer platforms

Appwrite is an open-source developer platform that provides backend services such as authentication, databases, storage, functions, messaging, and sites, and is described as an open-source cloud for agents and developers. It adopts an MCP and agent-first design, enabling AI agents to directly operate real-time projects, such as querying PostgreSQL, charting traffic, or publishing marketing campaigns. The platform centralizes the services needed to build applications in one place, covering modules such as user authentication, data modeling and querying, file storage, serverless function deployment, frontend hosting, messaging notifications, edge security, and real-time event subscriptions, and emphasizes a consistent experience for modular products from prototype to production scale. Appwrite is compatible with the frameworks, languages, and agents commonly used by developers, and continuously benchmarks performance against various leading models. On security, it has built-in DDoS protection, encryption of data at rest and in transit, abuse protection, data migration, GDPR support, and controls for sensitive data such as healthcare data. Its network integrates backend APIs, functions, and hosted websites, delivers services through a CDN and edge DDoS protection, and supports choosing global regions and edge nodes to optimize latency, compliance, and data residency, claiming latency of less than 50 milliseconds can be achieved. Currently, more than 500,000 developers use it, and official data mentions metrics such as the number of Discord members, monthly requests, and monthly database operations.

Read more

ImagineToVideo

AI video

ImagineToVideo is an AI image-to-video and text-to-video generator powered by models such as Seedance 2.5, Kling V3, and VEO 3.1. Registration is free, with no subscription, and no free credits; generating a video for the first time requires a one-time payment of $0.99 for 30 credits. It supports six creation modes: text-to-video, image-to-video, video-to-video, multi-scene, product video, and templates. It can generate MP4 videos with synchronized audio, paid plans are watermark-free, and it supports resolutions up to 4K.

Read more

ImagineX

AI video

ImagineX is an AI content generation platform, officially positioned as a professional AI video and image generation tool for creators, marketers, and enterprise users, focused on creating videos and images quickly, easily, and with high quality. Currently, public information provides only a one-sentence official introduction and does not disclose specific models, functional modules, workflows, pricing plans, platform support, or usage limitations; this information has not yet been provided.

Read more

imcodes

AI agents

IM.codes is an "instant messaging, memory, and review layer" for AI agents, open-sourced by GitHub user im4codes. It connects coding agents in the terminal to remote computers: supported machines can be registered as restricted "Controlled Nodes," allowing authorized agents to directly operate one or more computers with scope-constrained instructions, file transfers, and typed Computer Use, and when necessary, open a remote desktop takeover of a Windows node in a browser or on a phone. Beyond machine control, it also provides shared agent context and memory across providers, a managed MCP tool surface, supervised execution and automatic review, OpenSpec automatic delivery, collaborative coding, cross-provider team review, and other features, and can connect to Claude Code, Codex, Gemini CLI, GitHub Copilot, Cursor, OpenCode, OpenClaw, Qwen, and others. The project is positioned not as an AI IDE or a general-purpose remote terminal, but as the messaging, memory, and review layer around terminal coding agents. The author states that this is a personal open-source project, with no warranty, no SLA, and no guarantee of stability, security, or backward compatibility; use at your own risk.

Read more

imggpt

AI images

imggpt is a browser-based GPT Image image generation and photo editing tool. It integrates text-to-image generation, reference image/image-to-image control, one-click style effects, AI photo editing (background removal, object removal, portrait retouching, product image optimization), and export preparation into a single workspace, aimed at creators, marketers, e-commerce operators, designers, agencies, and teams for generating and refining visual assets such as ads, landing pages, social posts, blog covers, and product images.

Read more

ImgToVid AI

AI video

ImgToVid AI is an AI-driven creative platform whose core feature is an image-to-video generator that can transform photos, illustrations, characters, and product images into dynamic videos. The platform also integrates tools such as text-to-video, AI image generation, AI video editing, AI avatars, and creative effects, supports a variety of AI video models, and is suitable for social media content creators, marketers, designers, and storytellers. It offers free AI image and short video generation at basic quality, while higher resolution, longer duration, and advanced output require credits.

Read more

ImgVid

AI video

ImgVid is an AI image and video generation tool that lets users create images, edit photos, or turn text and images into short videos using text prompts or uploaded images. It offers six workflows: text-to-image, image-to-image, text-to-video, image-to-video, reference-image-to-video, and start-and-end-frame-to-video, and supports use in desktop or mobile browsers. Generated results are saved to the asset library within the account, where users can preview and download them, and complete branding, subtitles, or longer sequences in external editors. Free accounts can use free credits to generate watermarked content, and some models or features may require a paid plan or credit pack; default public visibility settings vary by account type. The tool does not provide standalone music or voice generation, generation time depends on factors such as model, queue, input, duration, and resolution, and prompts cannot precisely control details such as faces, hands, or text.

Read more