Minduck
AI imagesMinduck is a generating platform base on the changing of mind-map. Who are we : AI image generator ; AI writer ; AI mind
35,738 results · Page 304 of 993
Minduck is a generating platform base on the changing of mind-map. Who are we : AI image generator ; AI writer ; AI mind
Title: Create Unique Minecraft Skins with Stable Diffusion Model Meta Description: Discover how to generate custom Minecraft skins effortlessly using the Stable Diffusion model. Unleash your creativity and stand out in the game! Are you looking to personalize your Minecraft experience? With the Mine
Parse PDFs, images, DOCX and PPTX via the MinerU document parsing API, with multilingual OCR, batch processing, page ranges and local file upload.
Ming-omni-tts is a unified audio generation model launched by inclusionAI. It can jointly generate speech, ambient sound, and music within a single channel, and supports fine-grained control over speech rate, pitch, volume, emotion, dialect, and more. It uses a 12.5Hz continuous tokenizer and Patch-by-Patch compression to improve inference efficiency.
Google Chrome, use GPT-4o for free
DALL-E 3 is an AI image generator that creates detailed visuals from text prompts.
MINI_LLM is a personal practice project for experimenting with and reproducing the pretraining pipeline of a small-parameter Chinese large language model. The project's base model is QWEN, and a pretrained model with approximately 1.4B parameters was trained, with about 8B training tokens. The repository covers the complete pipeline of pretraining, SFT instruction fine-tuning, and DPO (Direct Preference Optimization), and includes DeepSpeed distributed training techniques; the PPO part is marked as to be done. The project provides download links for model weights at each stage of pretraining, SFT, and DPO, and uses SwanLab to view training logs. It is suitable for developers who want to learn or reproduce the Chinese large model training pipeline.
Mini_RWKV_7 is a small language model project trained on the RWKV-V7 architecture, with only 34.2M parameters and a model file of about 68.4MB. The project provides complete training code, datasets, and inference support, including three architecture variants: pure RWKV V7, RWKV V7 with deep embedding, and RWKV V7 with deep embedding attention. This model is suitable for deployment and rapid iterative development on devices with extremely limited resources, and can even be trained on a thin-and-light laptop with a Radeon 780M integrated GPU. The project has open-sourced the pretrained and SFT model weights, and provides an online experience via HuggingFace Space.
With the release of the GPT-4V multimodal model, ...
MinimalChat is a once-popular open-source AI communication tool built on Generative Pre-trained Transformers (GPT). It is characterized by being lightweight and easy to use, supporting file upload and import, multi-conversation management, history import and export, and featuring a guided operation flow and mobile-friendly compact navigation. The tool also provides voice interaction (speech-to-text STT, text-to-speech TTS), supports multiple language models (OpenAI-compatible APIs and local models), can be self-hosted via Docker, and supports running local models in the browser. Other features include swipe gestures for quick access to settings and conversations, Markdown and code syntax highlighting, mobile and PWA support, and it is open source under the MIT license. According to official statements, MinimalChat has officially ceased operations, and its philosophy of simplified and efficient AI communication is carried on by AI Chat, which is a comprehensive assistant platform integrating multiple mainstream AI models.
AI voice synthesis tool launched by MiniMax, supporting voice cloning
MiniMax Open Platform
MiniMind-V is an open-source multimodal vision-language model (VLM) project that provides complete code and tutorials for building a 65M-parameter ultra-small VLM from scratch, covering the full pipeline of dataset cleaning, Pretrain, and SFT, and enabling fast training and inference on a personal GPU.
A minimalistic AI-powered search engine that helps you find information on the internet.
MiniValuator is a stock valuation analysis tool built by a former auditor, helping users view the intrinsic value of US stocks and giving clear judgments based on companies' public financial reports. It does not predict stock prices or chase news; instead, it performs valuation and red flag checks according to fixed rules. Users can get their first analysis for free, and after saving a stock to "Second Read," the system will recheck it over time and explain the reasons in plain language when the judgment changes.
Personal finance management
MIOSN helps find the optimal LLM at the lowest cost for your tasks.
Unlike AI tools that stop at interview + transcript, Mira is a full AI researcher — plans studies, recruits globally (10…
AI platform for community management, image generation, chat, and search.
AI-powered platform for 3D design generation and asset search without tagging.
Comprehensive Wellness Platform That Puts Experts in Your Pocket 24/7
Easy-to-use online design tool with templates, graphics, and AI-powered features.
MiroFish is an AI prediction workflow tool that turns text, PDF, MD, and TXT files into graph building, simulation, reports, and follow-up conversations. Users can ask questions directly, and the system runs continuous workflows in the background such as seeding, simulation, and reporting, presenting summaries, report entry points, and follow-up paths as structured result cards. It is suited to scenarios that require simulating crowd reactions and narrative paths, such as pricing adjustments, brand ambassador changes, policy debates, market stories, and crisis response. The product emphasizes keeping structure, roles, social dynamics, and report synthesis within the same sequence for users to continue asking follow-up questions, positioned as exploratory decision support rather than precise prediction.
Mirro is a social styling app for outfit inspiration and AI-powered styling.
Mirror Emoji Keyboard by Mirror AI is an app that takes a photo of your face and creates you a set of personalized emoji. You get hundreds of them ready to be shared on messengers or social networks.
Historical third-party summary: The linked product is Fitbod, a fitness app described as organizing personalized strength workouts around goals, experience and available equipment. Product descriptions do not establish fitness outcomes.
AI-powered astrology platform for personalized insights and guidance.
Endel is a neuroscience-based sound health app that provides real-time personalized soundscapes to help users focus, relax, and fall asleep. Its patented technology dynamically adjusts sound based on inputs such as time, weather, heart rate, and location. Endel is available on iOS, Android, Apple Watch, Alexa, Apple TV, Fire TV, desktop, and web player, with cumulative downloads exceeding 4 million. It has received honors including Apple Watch App of the Year 2020, App Store Editor's Choice, and Google Play Best App 2021. Users can use sound scenes or preset scenarios to focus on work, relax, sleep, exercise, and more. The app includes binaural beat scenarios, and wearing stereo headphones is recommended for use. Endel uses a paid subscription model; specific pricing is not currently provided.
Gift Ideas AI is an AI gift planning tool designed around gift-giving scenarios, with its official website at https://giftideasai.com/。它不只是提示词式的 AI gift recommender, but a space that centrally manages the entire gift-giving process: you can save recipients, important dates, wishlists, gift cards, and reservation information, and use the AI gift finder and gift guides to provide direction when ideas are vague. The tool emphasizes privacy by default, supports magic link login, and offers guided setup on demand. It addresses gift-giving challenges with features such as a recipient library, date tracking, private surprise lists, shared family lists, wishlist reservations, gift detail records (price, files, expiration dates, receipts, notes), duplicate purchase prevention, and joint contributions, making it suitable for birthdays, holidays, weddings, anniversaries, and group gifting scenarios.
Miso One: Realistic AI Text-to-Speech Engine
Read-only MISP threat intelligence (events, attributes, search, tags, feeds, galaxies) with prompt injection scanning of every response.
📟 🏠 🐧 🍎 - Remote control for MiSTer-FPGA: launch games, search ROMs, take screenshots, manage systems and set up Tailscale VPN, with automatic core and ROM discovery.
Mistral is a learning and retention platform for organizing and retaining knowledge effectively.
A light-weight codebase that enables memory-efficient and performant finetuning of Mistral's models.
Blazingly fast LLM inference.
Mistress.to is an Immersive femdom AI companion featuring interactive audio voice calls, text chat, and high-quality image and video generation.