All AI tools

Nano Image AI is an online AI image editing and generation platform that offers text-to-image, image-to-image, photo editing, image upscaling and enhancement, and other features. It supports uploading common formats such as JPG, PNG, WEBP, and GIF (up to 10MB, up to 10 images). The platform claims that generated images are watermark-free and that copyright belongs to the user, and they can be used for personal or commercial projects. It also uses a zero-data-retention policy to process uploaded and generated content. New users can receive free credits and start using it without registration, while advanced use can be accessed through subscriptions or credit packs.

Nano Image AI

Nano Image AI is an online AI image generation and editing tool powered by the Google Gemini image model. It supports text-to-image generation, editing of existing photos, image upscaling and enhancement, and image-to-image conversion, all performed within a web interface. The tool is aimed at creators, marketers, and artists, and highlights 4K resolution support, in-image text rendering, privacy-first design (zero data storage), and commercial-use licensing.

Nano Pro

Nano Pro is a browser-based AI photo editing and image generation workspace that integrates Google's three image models—Nano Banana, Nano Banana Pro, and Nano Banana 2—into a single interface. Users can upload up to 14 reference photos and edit or generate with text descriptions, with support for 1K, 2K, and 4K output, and the fast model returns results in about 20 seconds. After registering, users receive 15 welcome credits, which can be used to generate one image with the fast model; free images have watermarks, and paid plans can remove watermarks. In addition to the Nano Banana series, the workspace also offers other image engines to choose from, including GPT Image 2, Seedream, and Qwen Image.

nano-aha-moment

nanoAhaMoment is a single-file, single-GPU, from-scratch implementation of an LLM reinforcement learning (RL) training library developed by McGill-NLP, implementing a DeepSeek R1-zero style training pipeline. It supports a single 80G GPU as well as multi-GPU training, covering 3B base models and 7B models, and uses full-parameter tuning. Its efficiency is competitive with verl, but its code is simpler. The project is inspired by TinyZero and Mini-R1 and is designed to be more concise, clear, and fast, with every line of code visible and easy to understand. The main files include the interactive Jupyter notebook with a tutorial, nano_r1.ipynb, the corresponding Python script, nano_r1_script.py, and a checkpoint comparison notebook. The goal is to provide researchers with an efficient and easy-to-understand RL for LLM training tool built from scratch.

Nano-Bananary

Nano-Bananary (Banana Supermarket) is an AI image and video generation tool based on Gemini 2.5 Flash Image, open-sourced on GitHub. It focuses on "no prompt needed" one-click playstyle generation, with a variety of built-in presets, allowing users to quickly generate images without writing prompts. The tool supports local brushing and selection, enabling editing of specific areas of an image; it supports continuous editing, where each generated output can be directly used as input for the next edit or generation, achieving iterative creation. It also provides a history feature, making it easy to call previously generated images as new input. The interface supports one-click switching between Chinese and light themes, and a video generation feature has been launched, allowing each playstyle to be converted into a video with one click. It also includes comparison viewing features: enlarged preview, horizontal comparison and comparison image download, slider comparison, and a custom prompt module. The project requires a Node.js environment, installs dependencies via npm, configures GEMINI_API_KEY in .env.local, and runs npm run dev to start. It can be used online in AI Studio.

nano-graphrag

A simple, easy-to-hack GraphRAG implementation.

nano-vllm

nano-vllm is a lightweight vLLM project implemented from scratch, aiming to provide offline inference speed comparable to vLLM while keeping the code concise and readable, with the entire implementation being about 1200 lines of Python code. It integrates optimization techniques such as prefix caching, tensor parallelism, Torch compilation, and CUDA graph, making it suitable for developers who need to understand the internal implementation of large model inference engines or perform lightweight deployment.

nano-world-model

Nano World Model (nano-world-model) is a minimalist, out-of-the-box open-source codebase for advancing World Model scientific research. It trains video world models based on the diffusion-forcing method, provides a unified pipeline from environment setup, data loading, training, validation to evaluation, and is managed through the Hydra configuration system. The project emphasizes scientific transparency, has clear code, supports controlled ablation experiments on prediction targets, action injection methods, and model scale, and fully open-sources model checkpoints. Its capabilities cover applications such as long-horizon autoregressive rollout, video-to-3D point cloud reconstruction, and MPC/CEM-based planning. The repository contains training, inference, and evaluation code, along with example scripts and detailed documentation, making it easy for researchers to quickly reproduce and extend.

nano_rl

nano_rl is a secondary development project based on verl, mainly used for custom development of reward (reward model) on the verl framework. The project aims to enhance the flexibility and performance of the original framework, with a focus on optimizing the invocation and management of reward models. According to the project materials, its content covers practical methods for custom reward models, as well as experience summaries from Agent training in verl.

NanoBanana 2 API

NanoBanana 2 API (Banana 2 API) is a developer-oriented AI image generation RESTful API service that calls the Banana 2 model through standard HTTP requests to generate high-quality, photorealistic images from text prompts. The official description states that a single image is typically generated within 10 seconds, supports resolutions from 512×512 to 4K and multiple aspect ratios, and supports both text-to-image and image-to-image (reference image style transfer, editing, variations). The usage flow is: register for free to obtain an API key, send a POST request to the endpoint with the prompt and parameters, receive a task ID, then poll the status endpoint to retrieve and download the image. It supports examples in Python, Node.js, JavaScript, cURL, and other languages and SDKs, uses Bearer Token authentication, HTTPS encryption, rate limiting, and GDPR-compliant data processing, and has automatic scaling capabilities. Pricing is a pay-as-you-go credit model, new accounts receive free credits with no credit card required, and generated images come with a commercial license. It is suitable for scenarios such as embedding image generation in SaaS products, batch production of e-commerce product images and marketing materials, marketing automation creatives, and game concept art and texture assets.

Nanobanana Pro

Nano Banana 2 & Pro is a suite of AI image generation and editing tools focused on generating and editing professional-grade 4K images. According to the official introduction, it excels at accurate text rendering, character and product visual consistency, and adherence to complex instructions, and can be used to generate labels, mathematical formulas, multilingual content, infographics, charts, product images, and marketing materials. The website provides a creation workspace where you can choose a model or tool to start creating, and supports reference image uploads. When using it, enter detailed prompts, select image dimensions and advanced options, then generate and download. Registration grants free credits for a trial. Paid plans are billed by credits, credits never expire, and include commercial usage rights, among other things; payments are processed by Stripe. The official statement says that due to computational costs and instant delivery, refunds are not supported.

NanoGPT

NanoGPT is a browser-based AI photo editing tool. Users only need to upload a photo and describe what they want to change using natural language text prompts to replace backgrounds, change clothing, remove or replace objects, change colors, and transform photo styles. It can also enhance, restore, and creatively transform generated results, making it suitable for creators, marketers, online sellers, and others who need quick targeted edits, but it is not suitable for manual retouching scenarios that require pixel-level precision.

nanoGRPO

nanoGRPO is a lightweight implementation of Group Relative Policy Optimization (GRPO), aimed at language model training. According to the repository description, users can use it to train their own "tiny r1/o1" models; the author provides results after training for 120 steps on an 8GB RTX 4060. Its core formula is: J_GRPO(θ) = (1/G) Σ_{i=1}^{G} (min(π_θ(o_i|q)/π_θ_old(o_i|q) · A_i, clip(π_θ(o_i|q)/π_θ_old(o_i|q), 1-ε, 1+ε) · A_i)). The repository points out that GRPO can be understood as PPO using the z score of grouped output as the advantage estimate, without using a value function, and provides links to the original paper and the r1 paper.

Nanoimagine

Nanoimagine is a unified multimodal AI generation API platform that provides image, video, and text generation capabilities through a single API key, a single billing system, and a single developer workflow. Developers can use the same set of interfaces to call models from multiple leading AI providers, submit asynchronous generation tasks, query task status, obtain result links, and switch between different models without rebuilding integrations. The platform supports image generation and editing, text-to-video and image-to-video, as well as OpenAI-compatible chat, agent, summarization, and structured text workflows, and provides features for production teams such as usage logs, project management, and API key access control. The models mentioned in the materials include GPT-Image-2, Nano Banana 2, Seedream 5.0 Pro, FLUX.2 Pro, Kling 3.0, xAI Grok Imagine 1.5 Preview, Claude Fable 5, Kimi K3, and others. Specific pricing, free quota, and SLA details are not yet provided.

NanoMaker

NanoMaker AI is an AI creation platform that integrates image, video, music, and audio generation. According to its official introduction, it brings together models such as Nano Banana 2, GPT image 2, and Seedance 2.5, providing access to multiple top-tier AI models through a single subscription. Users can describe editing requirements in natural language and upload images (optional), and the platform instantly generates results. It supports image generation, video generation, background replacement, lighting control, artistic style transfer, color modification, age changes, and other functions, and also provides generation capabilities such as character consistency and scene fusion. The platform includes a large gallery of preset prompts, covering scenarios such as LinkedIn headshots, 3D figurines, line art, memes, YouTube thumbnails, product photography, old photo restoration, and professional headshots, and is aimed at marketers, content creators, educators, social media operators, designers, e-commerce sellers, and other groups. The official claim is that one subscription provides access to all models, with no hidden fees and no separate per-model billing.

NanoPhoto.AI

NanoPhoto.AI is a professional AI video and photo editing platform, offering AI video editing (Sora 2 Pro, Sora Storyboard) and AI photo editing (Nano Banana) features. The platform supports creating longer, more controllable AI videos through multimodal reference materials, and provides production-oriented workflows. In addition, the platform also offers a Seedance 2.5 prompt exploration feature, where you can browse 1,152 creator prompts and their corresponding preview videos.

Nanorater

Nanorater is a free AI face rating and attractiveness test tool. After users upload a clear front-facing photo, they receive a face score from 0–100, the reasons behind the score, and one improvement suggestion to try. No registration is required for the first result. The tool uses a two-layer mechanism: an instant face map generated in the browser (about 1 second, measuring mirror balance, thirds, fifths, and aspect ratio; the photo is not uploaded and there is no usage limit); and scoring of the photo by a vision AI model, giving a 0–100 score based on six dimensions (facial harmony, symmetry and balance, skin quality, eyes and expression, lighting and color, styling and grooming), accompanied by visible evidence and a retake plan. The score explains how the photo performs, not a person's worth. A free account can unlock the full single-photo report, and there is also an optional one-time payment of $17.49 for the Glow-Up report, which includes an AI preview image after applying the improvements.

NanoSkill AI

AI agents

Discover and compare install-ready marketing agent skills for SEO, ads, email, and growth.

nanoVLM

nanoVLM is an open-source repository released by Hugging Face for training and fine-tuning small vision-language models (VLMs). It is implemented in pure PyTorch, with concise and readable code. The model definition and training logic total about 750 lines, including the vision backbone, language decoder, modality projection, and training loop. The default configuration uses SigLIP-B/16-224-85M and SmolLM2-135M as backbones to build a 222M-parameter model, which can be trained on a single H100 in about 6 hours on 1.7 million samples, reaching 35.3% accuracy on MMStar. The repository provides training, generation, and evaluation scripts, supports Hugging Face Hub integration and local save/load, and includes a VRAM measurement tool. It is suitable for researchers and developers who want to get started quickly and experiment with small VLMs.

nanowhale

nanowhale is a tiny language model project open-sourced by Hugging Face, trained from scratch using the DeepSeek-V4 architecture with approximately 110 million parameters. The repository contains all the code, configurations, and tokenizer needed for pretraining and fine-tuning. The model implements the full feature set of DeepSeek-V4 at a tiny scale, including Multi-head Latent Attention (MLA), Mixture of Experts (MoE), Hyper-Connections, and Multi-Token Prediction (MTP). It is suitable for research, learning, or experimenting with small-scale implementations of the DeepSeek architecture.

Nanu

Nanu helps parents find baby names by meaning and gender, powered by GPT.

Nara ai

AI homework assistant for math and physics, using text or photo input.

Naria ai

Naria is a platform for creating and sharing illustrated children's stories.

Narkis.ai

AI-powered personal photography studio for lifelike photos and videos.

Narrai

AI video

Narrai simplifies adding voiceovers to videos with AI-generated scripts and music.

NarraLand

AI office

AI-powered presentation content creation platform

Narralize

AI audio

Narralize converts PDFs into multilingual audio summaries using AI.

Narrato

AI writing

Narrato is an AI writing and content marketing platform for enterprise marketing teams. According to available information, it emphasizes the ability to orchestrate people, AI, and systems in an agentic manner to produce personalized marketing campaign content that matches brand tone at scale. The platform treats brand guidelines, audience information, material assets, and marketing experience as a structured intelligence foundation for AI agents to execute and follow, thereby reducing chaos in marketing campaigns and improving consistency and measurable marketing results. Users can describe their needs in natural language, and agents plan, execute, self-check, and iterate until preset standards are met; real performance data from each channel also flows back as new experience, making subsequent marketing campaigns smarter. Narrato can work with commonly used tools such as Figma, Photoshop, and Illustrator, and supports integration with existing ecosystems through MCP, APIs, and integrations. The platform also provides security and governance controls, emphasizing brand safety, data compliance, and access control. It should be noted that the available information does not provide specific pricing, plans, free trial policies, supported platform details, or detailed onboarding steps; this information is not currently available.

nashnova

AI agents

An AI investment research workstation for individual investors, covering A-shares, Hong Kong stocks, and U.S. stocks, with more than twenty built-in strategies and nearly a hundred expert avatars

Nastia ai

AI assistants

Uncensored AI companion for chat, roleplay, and emotional support.

NationBuilder

AI marketing

NationBuilder is a people-centered technology platform for nonprofits, causes, campaigns, and social movements. It provides native advocacy features, a dynamic people database, payment processing and fundraising, hosted action websites, and email and SMS outreach tools, designed to help organizations grow their communities and drive action. The platform is suitable for a variety of communities and organizations, regardless of goal, and can be used to run a nonprofit, advance advocacy, win elections, or build a brand, and claims you can get started in 15 minutes or less.

NativeShip

BillionVerify is an email verification tool built for scale, offering real-time SMTP validation, bulk list cleaning, and an API, and claiming 99.9% accuracy. It verifies email addresses through multiple checks including syntax, domain, MX, SMTP, reputation, and risk, supports one-by-one verification, bulk file submission, real-time API calls, and the ability to build a white-label verification portal. The product is aimed at teams such as sales, marketing, and B2B agencies that need to clean outreach lists and prospect lists, offers 20 free credits per day (up to 600 per month), supports MCP server integration, and requires no credit card for the free tier. Note: the materials provided by the user do not mention specific pricing, a list of integration tools, or other details.

NativeShip