SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,708 results · Page 873 of 992

sourcebook

AI coding

sourcebook is a code change completeness detection tool for AI coding agents, with the core goal of discovering related files that AI agents "forget to change" when modifying code, acting as a safety layer for code changes. By analyzing git diff, combined with the repository's real structure and git history, it flags files that should have been modified together but were not, such as test files not synchronized after source file changes, sibling modules that import or depend on the file, "buddy files" that have historically often changed together, and the scope of impact involving hub files with many dependents. The tool uses a two-layer mechanism: Layer A is rule-based, does not call large models, and runs in under one second, including co-change analysis, test file detection, import graph construction, and hub detection; Layer B uses AI for semantic analysis, which can catch deep issues such as field renames requiring migration, outdated validation logic, and cross-module dependencies without import relationships. All AI suggestions must provide dependency references, and hallucinated paths are filtered, ensuring zero false positives on clean diffs. In terms of usage, it can be run directly on any diff without configuration; it supports integration into Claude Code's pre-commit hook with a single command; it has been published to the official MCP registry, allowing agents to query repository structure, impact scope, conventions, and co-change data on demand; the GitHub App is in a coming soon state with a waitlist open. The project is built on real benchmarks, which include 19 tasks, 10 repositories, and 4 languages. It uses the BSL-1.1 source-available license, can be used for free but cannot be offered as a hosted service, and converts to the MIT license on March 25, 2030.

Read more

Soutine AI

AI design

Soutine AI is an all-in-one AI creation workbench for image and video generation, integrating multiple leading models into a single space. Users can generate content via prompts or reference images, and all results are saved in a shared history. The official page shows supported models including Kling 4, Qwen Image 2.1, Nano Banana, Nano Banana 2 Lite, Nano Banana Pro, GPT Image 2.5, and others, which can be used for text-to-image, reference-image editing, image-to-video, AI video upscaling to 4K, and more. It offers a free starting quota, and paid subscriptions are divided into three tiers: Starter, Pro, and Max, supporting commercial use and watermark-free export. Generation tasks display credit consumption before submission, billed by model, resolution, duration, and input type.

Read more

SoVideo

AI video

SoVideo is an AI video generation tool that can quickly turn text or images into high-definition, ready-to-use dynamic videos. It supports text-to-video and image-to-video, and can automatically align audio to achieve natural rhythm and smoothness. Users only need to upload a character image and provide a reference video to generate and download a video in the cloud. The platform offers multi-aspect-ratio support, AI quality enhancement, watermark-free export, and character consistency, making it suitable for directors, content studios, brand marketing, education, and product teams.

Read more

Space Bunny

AI developer platforms

Space Bunny is an anonymous long-context reasoning model offering a 1 million token context window, supporting text, image, and video inputs, with adjustable reasoning effort (low, medium, high, xhigh, max). It is aimed at developers, providing an OpenAI-compatible REST API that supports the standard chat completions format, streaming output, JSON responses, and tool calling, with the application holding the API key on the server side and retaining control over tool-calling permissions and side effects. The model can be used for code review, long-document reasoning, visual understanding, and agentic workflows. Space Bunny is currently in preview, and its model architecture, parameter scale, training details, and knowledge cutoff are all undisclosed. Usage is based on a usage-based credit system, with credits valid permanently, purchased one-time, with no subscription or automatic renewal, divided into tiers such as 100K, 1M, and 11M credits, with different tiers differing in the number of concurrent requests, number of workspaces, and support level, and it can be tried for free in the online playground.

Read more

Spanish TTS

AI audio

Spanish TTS is a free online Spanish text-to-speech tool that converts Spanish text into natural, realistic AI speech without registration. It offers multiple accents including Castilian (Spain), Mexican, Latin American, and US Spanish, with dozens of built-in male and female voices in different styles, and can generate studio-quality Spanish audio that can be downloaded as MP3. It can also be paired with translation tools to achieve English-to-Spanish dubbing, making it suitable for scenarios such as video narration, pronunciation practice, reading articles aloud, and story recording.

Read more

SparkIconAI

AI design

SparkIconAI is a professional app icon generator for iOS, Android, Web, and desktop, primarily serving mobile developers and indie developers. Users can describe an app concept in natural language or provide a reference image to generate a 1024x1024 primary icon, and preview the result in real time under real platform masks. The platform supports simulated checks for scenarios including iOS continuous-curvature rounded corners, Android adaptive masks (circle, squircle, rounded rectangle, teardrop), macOS Dock reflections, and browser 16px/32px tabs. After generation, users can export a complete asset package in one click, including Xcode Asset Catalog, Android mipmap, Web/PWA icons, and more. The product emphasizes node-based infinite canvas iteration, one-click background removal with transparent/solid color switching, and safe-area padding adjustment, and claims that all image processing is done locally in the browser.

Read more

SparkVid

AI video

SparkVid is an online AI video generator that turns text descriptions or static images into videos. It integrates multiple AI video models such as Seedance, Kling, and Veo within a single browser page, supporting text-to-video, image-to-video, motion control, reference-video motion transfer, and video editing via natural language descriptions, and it can generate starting frames with GPT Image, Nano Banana, and others. Users do not need to install software or a GPU and can use it in desktop and mobile browsers. After generation, they can preview, adjust, and export standard MP4, with support for aspect ratios such as 16:9, 9:16, and 1:1 and resolutions up to 4K.

Read more